du.sePublikationer
Ändra sökning
Avgränsa sökresultatet
123 101 - 145 av 145
RefereraExporteraLänk till träfflistan
Permanent länk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • chicago-author-date
  • chicago-note-bibliography
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Träffar per sida
  • 5
  • 10
  • 20
  • 50
  • 100
  • 250
Sortering
  • Standard (Relevans)
  • Författare A-Ö
  • Författare Ö-A
  • Titel A-Ö
  • Titel Ö-A
  • Publikationstyp A-Ö
  • Publikationstyp Ö-A
  • Äldst först
  • Nyast först
  • Skapad (Äldst först)
  • Skapad (Nyast först)
  • Senast uppdaterad (Äldst först)
  • Senast uppdaterad (Nyast först)
  • Disputationsdatum (tidigaste först)
  • Disputationsdatum (senaste först)
  • Standard (Relevans)
  • Författare A-Ö
  • Författare Ö-A
  • Titel A-Ö
  • Titel Ö-A
  • Publikationstyp A-Ö
  • Publikationstyp Ö-A
  • Äldst först
  • Nyast först
  • Skapad (Äldst först)
  • Skapad (Nyast först)
  • Senast uppdaterad (Äldst först)
  • Senast uppdaterad (Nyast först)
  • Disputationsdatum (tidigaste först)
  • Disputationsdatum (senaste först)
Markera
Maxantalet träffar du kan exportera från sökgränssnittet är 250. Vid större uttag använd dig av utsökningar.
  • 101.
    Paidi, Vijay
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Fleyeh, Hasan
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    Nyberg, Roger G.
    Högskolan Dalarna, Akademin Industri och samhälle, Informatik.
    Deep learning-based vehicle occupancy detection in an open parking lot using thermal camera2019Ingår i: Artikel i tidskrift (Övrigt vetenskapligt)
    Abstract [en]

    Parking vehicle is a daunting task and a common problem in many cities around the globe. The search for parking space leads to congestion, frustration and increased air pollution. Information of a vacant parking space would facilitate to reduce congestion and subsequent air pollution. Therefore, aim of the paper is to acquire vehicle occupancy in an open parking lot using deep learning. Thermal camera was used to collect the data during varying environmental conditions such as; sunny, dusk, dawn, dark and snowy conditions. Vehicle detection with deep learning was implemented where image classification and object localization were performed for multi object detection. The dataset consists of 527 images which were manually labelled as there were no pre-labelled thermal images available. Multiple deep learning networks such as Yolo, ReNet18, ResNet50 and GoogleNet with varying layers and architectures were evaluated on vehicle detection. Yolo, GoogleNet and ResNet18 are computationally efficient detectors which took less processing time while Resnet50 produced better detection results compared to other detectors. However, ResNet18 also produced minimal miss rates and is suitable for real time vehicle detection. The detected results were compared with a template of parking spaces and IoU value is used to identify vehicle occupancy information.

  • 102.
    Paidi, Vijay
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Håkansson, Johan
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Nyberg, Roger G.
    Högskolan Dalarna, Akademin Industri och samhälle, Informatik.
    A holistic decision support system for last mile handovers2019Ingår i: Artikel i tidskrift (Övrigt vetenskapligt)
    Abstract [en]

    The last mile handover is assumed to be the most problematic part in the delivery process and the costs can go upto 50% of the total logistic cost. Real time consumer communication and dynamic scheduling are the major problem areas associated with effective attended last mile handovers. Therefore, aim of this paper is to report the design and development of a holistic decision support system’s functionalities which simultaneously addresses real time consumer communication and dynamic scheduling. A decision support system was designed and developed based on workshops, expert group interviews and its functionalities were proposed with the use cases. A survey was conducted with consumers of a retailer where majority of the consumers accepted the use of mobile communication devices to enable real time communication and alternate handover location which improves customer satisfaction and facilitates to avoid missed handovers. A pilot test was performed where routing distance was reduced with the use of optimized handover routes. However the improvement is subjected to the experience of driver and real time traffic conditions. We conclude that a holistic decision support system with multi-party communication among the stakeholders facilitates in reducing operational costs for logistic companies and improving customer satisfaction and business opportunities.

  • 103. Pan, S
    et al.
    Xiong, Y
    Han, Y
    Zhang, Xingxing
    Högskolan Dalarna, Akademin Industri och samhälle, Energiteknik.
    Xia, L
    Wei, S
    Wu, J
    Han, Mengjie
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    A study on influential factors of occupant window-opening behavior in an office building in China2018Ingår i: Building and Environment, ISSN 0360-1323, E-ISSN 1873-684X, Vol. 133, s. 41-50Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Occupants often perform many types of behavior in buildings to adjust the indoor thermal environment. In these types, opening/closing the windows, often regarded as window-opening behavior, is more commonly observed because of its convenience. It not only improves indoor air quality to satisfy occupants' requirement for indoor thermal comfort but also influences building energy consumption. To learn more about potential factors having effects on occupants' window-opening behavior, a field study was carried out in an office building within a university in Beijing. Window state (open/closed) for a total of 5 windows in 5 offices on the second floor in 285 days (9.5 months) were recorded daily. Potential factors, categorized as environmental and non-environmental ones, were subsequently identified with their impact on window-opening behavior through logistic regression and Pearson correlation approaches. The analytical results show that occupants' window-opening behavior is more strongly correlated to environmental factors, such as indoor and outdoor air temperatures, wind speed, relative humidity, outdoor FM2.5 concentrations, solar radiation, sunshine hours, in which air temperatures dominate the influence. While the non-environmental factors, i.e. seasonal change, time of day and personal preference, also affects the patterns of window-opening probability. This paper provides solid field data on occupant window opening behavior in China, with high resolutions and demonstrates the way in analyzing and predicting the probability of window-opening behavior. Its discussion into the potential impact factors shall be useful for further investigation of the relationship between building energy consumption and window-opening behavior.

  • 104.
    Pourghadiri Esfahani, Mohammad
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Effect analysis of the 2008 charge increase of the nitrogen oxide on "Adoptive NOx Intensity" for Swedish pulp and paper industry2015Självständigt arbete på avancerad nivå (masterexamen), 10 poäng / 15 hpStudentuppsats (Examensarbete)
    Abstract [en]

    The aim of this study is to evaluate the effectiveness of the NOx charge increase in 2008 on the behavior of NOx emissions from pulp and paper combustion plants, which are in the fee and refund system (regulated plants) in Sweden. In order to evaluate the potential change in NOx emissions behavior, this study used Difference-in-Differences (DID) approach. This involves a ‘control group’ (unregulated plants) and a ‘treatment group’ (regulated plants). The control group consists of smaller plants which are not included in the fee and refund system. The crudest DID assumption is that the average difference in emissions would have stayed constant over time for the control group and the treatment group if the increase of the fee would not have taken place. The results suggest that the NOx charge increase has had negative effects on NOx emission efficiency ("Adoptive NOx Intensity") for some years after 2008. However the reduction in NOx emission efficiency is inconsistent for the period over 2008-13. The difference between the control group and the treatment group jumps up and down in a way that is difficult to explain by the increase of the fee in 2008. This indicates that the DID assumption for the DID estimator may not be fulfilled. It is concluded that the results given these conditions are not reliable to draw any strong conclusions about the possible effect of the increase of the fee in 2008.

  • 105.
    Prochazka, Jiri
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Zerin, Sonia
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Modeling Hospital Readmissions in Dalarna County, Sweden2016Självständigt arbete på avancerad nivå (masterexamen), 20 poäng / 30 hpStudentuppsats (Examensarbete)
    Abstract [en]

    Unplanned hospital readmissions increase health and medical care costs and indicate lower the lower quality of the healthcare services. Hence, predicting patients at risk to be readmitted is of interest. Using administrative data of patients being treated in the medical centers and hospitals in the Dalarna County, Sweden, during 2008 – 2016 two risk prediction models of hospital readmission are built. The first model relies on the logistic regression (LR) approach, predicts correctly 2,648 out of 3,392 observed readmission in the test dataset, reaching a c-statistics of 0.69. The second model is built using random forests (RF) algorithm; correctly predicts 2,183 readmission (out of 3,366) and 13,198 non-readmission events (out of 18,982). The discriminating ability of the best performing RF model (c-statistic 0.60) is comparable to that of the logistic model. Although the discriminating ability of both LR and RF risk prediction models is relatively modest, still these models are capable to identify patients running high risk of hospital readmission. These patients can then be targeted with specific interventions, in order to prevent the readmission, improve patients’ quality of life and reduce health and medical care costs.

  • 106.
    Rahman, Md Rezaur
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Joy, Frank
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Estimation of the causal effect of hospital outlier on patient outcomes: A case study of the hospital and patient care units inDalarna County, Sweden2018Självständigt arbete på avancerad nivå (masterexamen), 10 poäng / 15 hpStudentuppsats (Examensarbete)
    Abstract [en]

    An outlier patient is a medical or surgical patient who cannot get admitted to the designated ward or care unit due to the lack of bed occupancy or human resources, and the hospital transfers the patient to another ward or care unit. This study aims to compare and evaluate the outcome of being an outlier patient with a non-outlier patient. Region Dalarna (Lanstinget Dalarna) in Sweden provided data of two hospitals with 158734 cases for the years from 2014 to 2017. This observational casecontrol study used three types of matching techniques to create balanced data sets. Multivariate analysis with logistic regression was used to analyze the outcome of patients regarding mortality rate and unplanned readmission rate. Multiple linear regression was used to perform outcome analysis for the hospital length of stay of the outlier patients. Fisher’s exact test was used to evaluate the significance of mortality rate and unplanned readmission in 30 days. For the patient’s length of stay, the study used two independent t-test. Medical outlier patients did not get affected regarding unplanned readmission. In case of mortality, two of our matched datasets showed outlier patients did not have different mortality rate than non-outlier patients; only one matched dataset showed significance for mortality in case of outlier patients compared to the non-outlier patients. However, outlier patients had a significantly shorter duration of hospital stay than non-outlier patients in all three matched dataset.

  • 107.
    Raihana, Nishat
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Optimization of Operational Overhead based on the Evaluation of Current Snow Maintenance System: A Case Study of Borlänge, Sweden2019Självständigt arbete på avancerad nivå (masterexamen), 20 poäng / 30 hpStudentuppsats (Examensarbete)
    Abstract [en]

    This study analyzes snow maintenance data of Borlänge municipality of Sweden based on the data of 2017 to 2018. The goal of this study is to reduce operational overhead of snow maintenance, for example, fuel and time consumption of the snow maintenance vehicles, work hour of dedicated personnel, etc. Borlänge Energy equipped the snow maintenance vehicles with GPS devices which stored the record of the snow maintenance activities. The initial part of this study obtained insights out of the GPS data by using spatiotemporal data analysis. Derivation of the different snow maintenance treatments (plowing, sanding and salting) as well as the efficiency of the sub-contractors (companies which are responsible for snow maintenance) and inspectors (personnel who are liable to call the subcontractors if they think it is time for snow maintenance) are performed in the beginning of this study. The efficiency of the subcontractors and inspectors are measured to compare their performance with each other.

    The latter part of this study discusses a simulated annealing-based heuristics technique to find out optimal location for dispatching snow maintenance vehicles. In the existing system of snow maintenance, drivers of the maintenance vehicles decide to start location of maintenance work based on their experience and intuition, which might vary from one driver to another driver. The vehicle dispatch locations are calculated based on the availability of the vehicles. For example, if a subcontractor has three vehicles to perform snow maintenance on a specific road map, the proposed solution would suggest three locations to dispatch those vehicles.

    The purpose of finding the optimal dispatch location is to reduce the total travel distance of the maintenance vehicles, which yield less fuel and time consumption. The study result shows the average travel distance for 1, 3, and 5 vehicles on 15 road networks. The proposed solution would yield 18% less travel than the existing system of snow maintenance.

  • 108.
    Rajasekaran, Prabakaran
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    A statistical data analysis approach to Energy Data: A Case Study in Building Performance Analysis of Thermal Energy Loss2015Självständigt arbete på avancerad nivå (masterexamen), 10 poäng / 15 hpStudentuppsats (Examensarbete)
    Abstract [en]

    The usage of energy in buildings is higher in Sweden’s cold climate, most buildings consume

    significant energy to heat buildings during the winter and cool the buildings during the summer, using

    the district heat and electricity. The building energy loss is the difference between indoor and outdoor

    temperatures. When the temperature difference is higher during the heating season (winter), there is

    a need to balance the indoor temperature. More power supply is needed to warm the indoor area. In

    order to find which factor has a significant impact on the building, heat loss and gain can be used as

    variables in the multiple linear regression model to analyze the building energy performance. In order

    to build the multiple – linear regression model which is used, the measured parameters for the building

    and some of the data should be calculated, such as solar heat gain through windows from the (North,

    East, West &South) building. The heat loss to the ground is based on constructed material, the thermal

    conductivity of the material, indoor and outdoor temperature, and steady state ground heat transfer

    coefficient. After building the model, an analysis of the fit model test is needed, to investigate if the

    coefficients are properly estimated. Based on this analysis, we can see the comparison between

    renovated building and non renovated building significant impact of the energy consumption for

    given the energy and financial investment.

  • 109.
    Rebreyend, Pascal
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    Lemarchand, Laurent
    Massé, Damien
    Håkansson, Johan
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Multiobjective Optimization for Multimode Transportation Problems2018Ingår i: Advances in Operations Research, ISSN 1687-9147, E-ISSN 1687-9155, artikel-id 8720643Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    We propose modelling for a facilities localization problem in the context of multimode transportation. The applicative goal is to locate service facilities such as schools or hospitals while optimizing the different transportation modes to these facilities. We formalize the School Problem and solve it first exactly using an adapted -constraint multiobjective method. Because of the size of the instances considered, we have also explored the use of heuristic methods based on evolutionary multiobjective frameworks, namely, NSGA2 and a modified version of PAES. Those methods are mixed with an original local search technique to provide better results. Numerical comparisons of solutions sets quality are made using the hypervolume metric. Based on the results for test-cases that can be solved exactly, efficient implementation for PAES and NSGA2 allows execution times comparison for large instances. Results show good performances for the heuristic approaches as compared to the exact algorithm for small test-cases. Approximate methods present a scalable behavior on largest problem instances. A master/slave parallelization scheme also helps to reduce execution times significantly for the modified PAES approach.

  • 110.
    Rudholm, Niklas
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Nationalekonomi. HUI Research.
    Li, Yujiao
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Carling, Kenneth
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    How Does Big-Box Entry Affect Labor Productivity in Durable Goods Retailing? A Synthetic Control Approach. How Does Big-Box Entry Affect Labor Productivity in Durable Goods Retailing? A Synthetic Control Approach2018Rapport (Övrigt vetenskapligt)
    Abstract [en]

    Using  data  from  2001–2012,  the  effects  of  IKEA  entry  in  four  Swedish  municipalities,   2004–2007,   on   labor   productivity   in   durable   goods   retailing   is   investigated  using  synthetic  control  methods.  We contribute  to  the  literature  on  synthetic   control   methods   by considering   parametric   specifications   of   the   intervention effect, which in our case arguably improves the likelihood of identifying the intervention effect of IKEA entry on labor productivity. As inference relies on a single  treated  observational  unit  (i.e.,  a  single  IKEA-entry  municipality),  statistical  testing  is  a  challenge,  and  randomization  and  replication  for  inference  is  done  with  regard to the pool of control municipalities. Our results indicate that in three out of four  entry  municipalities,  labor  productivity  increased  more  than  in  their  synthetic  counterparts after IKEA entry, and that the size of the positive effect is related to the size of the new IKEA relative to the size of the existing durable goods retail sector in the entry municipalities, with larger positive effects found in municipalities where the new IKEA was large relative to the existing durable goods retail market.

  • 111.
    Rybarczyk, Yves
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys. Universidad de Las Américas, Quito, Ecuador.
    Carvalho, D. G.
    Cordella, F.
    Bioinspired Implementation and Assessment of a Remote-Controlled Robot2019Ingår i: Applied Bionics and Biomechanics, ISSN 1176-2322, E-ISSN 1754-2103, artikel-id 8575607Artikel i tidskrift (Refereegranskat)
  • 112.
    Sabah Almusleh, Azhar
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Probabilistic forecasting for one day ahead solar energy: Using linear and non-linear methods to forecast one day ahead solar energy output for photovoltaic power plant in Västerås, Sweden.2018Självständigt arbete på avancerad nivå (masterexamen), 10 poäng / 15 hpStudentuppsats (Examensarbete)
    Abstract [en]

    Photovoltaic (PV) systems become an important technology for generating electric power, but their output is not as stable and consistent as energy output from traditional fossil fuels. To help grow and integrate these new technologies into larger power systems and power grids, methodologies need to be improved to accurately forecast PV output.

    One particularly important type of forecast is One-Day-Ahead forecasting, and it is the focus of this paper. Factors such as solar irradiation, air temperature, and relative humidity can influence the output of PV systems, and the accuracy of the forecasting relies heavily on the chosen input variables. The research associated with this paper tests the hypothesis that, given the availability of weather forecast variables, these variables can be used instead of instant meteorological variables measured onsite (local variables) in predicting One-Day-Ahead (from sunrise to sunset) solar energy.

    This research was conducted to forecast the power production of an 8.4 kW solar site located at Västerås. Research and findings are reviewed, focused on two major techniques – linear, using Multiple Linear Regression (MLR), and non-linear, using Support Vector Regression (SVM). Both approaches are explored relative to developing the best One-Day-Ahead (sunrise to sunset) solar energy output forecast.

    The research utilized solar energy output data obtained from Västerås solar site as well as metrological variables from Copernicus Atmosphere Monitoring Service (GMES) for the period 01/05/2014 – 10/12/2015. Stepwise techniques were used to identify the most significant input variables as well as the application of a" sliding windows" technique for dataset construction (with varying historical time parameters).

    The results indicate a 30-day historical horizon provides the most accurate forecasts. Even though the research also showed that SVM achieved 16,4% more accurate predictions than MLR, Mean Absolute Percentage Error (MAPE) analysis indicates that SVM can only adequately predict (with MAPE below 30%) just over 50% of the time (188 days) over the course of a year. Thus, there are opportunities for future expansion and improvement to the research and methodologies described in this paper.

  • 113.
    Saqlain, Murshid
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Alam, Moudud
    Högskolan Dalarna, Akademin Industri och samhälle, Statistik.
    Brandt, Daniel
    Högskolan Dalarna, Akademin Industri och samhälle, Kulturgeografi.
    Rönnegård, Lars
    Högskolan Dalarna, Akademin Industri och samhälle, Statistik.
    Westin, Jerker
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    Stochastic differential equations modelling of levodopa concentration in patients with Parkinson's disease2018Konferensbidrag (Övrigt vetenskapligt)
    Abstract [en]

    The purpose of this study is to investigate a pharmacokinetic model of levodopa concentration in patients with Parkinson's disease by introducing stochasticity so that inter-individual variability may be separated into measurement and system noise. It also aims to investigate whether the stochastic differential equations (SDE) model provide better fits than its ordinary differential equations (ODE) counterpart, by using a real data set. Westin et al. developed a pharmacokinetic-pharmacodynamic model for duodenal levodopa infusion described by four ODEs, the first three of which define the pharmacokinetic model. In this study, system noise variables are added to the aforementioned first three equations through a standard Wiener process, also known as Brownian motion. The R package PSM for mixed-effects models is used on data from previous studies for modelling levodopa concentration and parameter estimation. First, the diffusion scale parameter, σ, and bioavailability are estimated with the SDE model. Second, σ is fixed to integer values between 1 and 5, and bioavailability is estimated. Cross-validation is performed to determine whether the SDE based model explains the observed data better or not by comparingthe average root mean squared errors (RMSE) of predicted levodopa concentration. Both ODE and SDE models estimated bioavailability to be about 88%. The SDE model converged at different values of σ that were signicantly different from zero while estimating bioavailability to be about 88%. The average RMSE for the ODE model wasfound to be 0.2980, and the lowest average RMSE for the SDE model was 0.2748 when σ was xed to 4. Both models estimated similar values for bioavailability, and the non-zero σ estimate implies that the inter-individual variability may be separated. However, the improvement in the predictive performance of the SDE model turned out to be rather small, compared to the ODE model.

  • 114.
    Saqlain, Murshid
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Alam, Moudud
    Högskolan Dalarna, Akademin Industri och samhälle, Statistik.
    Rönnegård, Lars
    Högskolan Dalarna, Akademin Industri och samhälle, Statistik.
    Westin, Jerker
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    Investigating Stochastic Differential Equations Modelling for Levodopa Infusion in Patients with Parkinson's Disease2019Ingår i: European journal of drug metabolism and pharmacokinetics, ISSN 0378-7966, E-ISSN 2107-0180Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    BACKGROUND AND OBJECTIVES: Levodopa concentration in patients with Parkinson's disease is frequently modelled with ordinary differential equations (ODEs). Here, we investigate a pharmacokinetic model of plasma levodopa concentration in patients with Parkinson's disease by introducing stochasticity to separate the intra-individual variability into measurement and system noise, and to account for auto-correlated errors. We also investigate whether the induced stochasticity provides a better fit than the ODE approach.

    METHODS: In this study, a system noise variable is added to the pharmacokinetic model for duodenal levodopa/carbidopa gel (LCIG) infusion described by three ODEs through a standard Wiener process, leading to a stochastic differential equations (SDE) model. The R package population stochastic modelling (PSM) was used for model fitting with data from previous studies for modelling plasma levodopa concentration and parameter estimation. First, the diffusion scale parameter (σw), measurement noise variance, and bioavailability are estimated with the SDE model. Second, σw is fixed to certain values from 0 to 1 and bioavailability is estimated. Cross-validation was performed to compare the average root mean square errors (RMSE) of predicted plasma levodopa concentration.

    RESULTS: Both the ODE and the SDE models estimated bioavailability to be approximately 75%. The SDE model converged at different values of σw that were significantly different from zero. The average RMSE for the ODE model was 0.313, and the lowest average RMSE for the SDE model was 0.297 when σw was fixed to 0.9, and these two values are significantly different.

    CONCLUSIONS: The SDE model provided a better fit for LCIG plasma levodopa concentration by approximately 5.5% in terms of mean percentage change of RMSE.

  • 115.
    Svenson, Kristin
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    A Microdata Analysis Approach to Transport Infrastructure Maintenance2017Doktorsavhandling, sammanläggning (Övrigt vetenskapligt)
    Abstract [en]

    Maintenance of transport infrastructure assets is widely advocated as the key in minimizing current and future costs of the transportation network. While effective maintenance decisions are often a result of engineering skills and practical knowledge, efficient decisions must also account for the net result over an asset's life-cycle. One essential aspect in the long term perspective of transport infrastructure maintenance is to proactively estimate maintenance needs. In dealing with immediate maintenance actions, support tools that can prioritize potential maintenance candidates are important to obtain an efficient maintenance strategy.

    This dissertation consists of five individual research papers presenting a microdata analysis approach to transport infrastructure maintenance. Microdata analysis is a multidisciplinary field in which large quantities of data is collected, analyzed, and interpreted to improve decision-making. Increased access to transport infrastructure data enables a deeper understanding of causal effects and a possibility to make predictions of future outcomes. The microdata analysis approach covers the complete process from data collection to actual decisions and is therefore well suited for the task of improving efficiency in transport infrastructure maintenance.

    Statistical modeling was the selected analysis method in this dissertation and provided solutions to the different problems presented in each of the five papers. In Paper I, a time-to-event model was used to estimate remaining road pavement lifetimes in Sweden. In Paper II, an extension of the model in Paper I assessed the impact of latent variables on road lifetimes; displaying the sections in a road network that are weaker due to e.g. subsoil conditions or undetected heavy traffic. The study in Paper III incorporated a probabilistic parametric distribution as a representation of road lifetimes into an equation for the marginal cost of road wear. Differentiated road wear marginal costs for heavy and light vehicles are an important information basis for decisions regarding vehicle miles traveled (VMT) taxation policies.

    In Paper IV, a distribution based clustering method was used to distinguish between road segments that are deteriorating and road segments that have a stationary road condition. Within railway networks, temporary speed restrictions are often imposed because of maintenance and must be addressed in order to keep punctuality. The study in Paper V evaluated the empirical effect on running time of speed restrictions on a Norwegian railway line using a generalized linear mixed model.

  • 116.
    Svenson, Kristin
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    McRobbie, S.
    Alam, Moudud
    Högskolan Dalarna, Akademin Industri och samhälle, Statistik.
    Detecting road pavement deterioration with finite mixture models2019Ingår i: The international journal of pavement engineering, ISSN 1029-8436, E-ISSN 1477-268X, Vol. 20, nr 4, s. 458-465Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Budget restrictions often limit the number of possible maintenance activities in a road network each year. To effectively allocate resources, the rate of road pavement deterioration is of great importance. If two maintenance candidates have an equivalent condition, it is reasonable to maintain the segment with the highest deterioration rate first. To identify such segments, finite mixture models were applied to road condition data from a part of the M4 highway in England. Assuming that data originates from two different normal distributions – defined as a ‘change’ distribution and an ‘unchanged’ distribution – all road segments were classified into one of the groups. Comparisons with known measurement errors and maintenance records showed that segments in the unchanged group had a stationary road condition. Segments classified into the change group showed either a rapid deterioration, improvement in condition because of previous maintenance or unusual measurement errors. Together with additional information from maintenance records, finite mixture models can identify segments with the most rapid deterioration rate, and contribute to more efficient maintenance decisions.

  • 117.
    Thomas, Ilias
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Automating levodopa dosing schedules for Parkinson’s disease2019Doktorsavhandling, sammanläggning (Övrigt vetenskapligt)
    Abstract [en]

    Parkinson’s disease (PD) is the second most common neurodegenerative disease. Levodopa is mainly used to manage the motor symptoms of PD. However, disease progression and long-term use of levodopa cause reduced medication efficacy and side effects. When that happens, precise individualized dosing schedules are required.

    This doctoral thesis in the field of Micro-data analysis introduces an end-to-end solution for the automation of the pharmacological management of PD with levodopa, and offers some insight on levodopa pharmacodynamics. For that purpose, an algorithm that derives objective ratings for the patients’ motor function through wearable sensors is introduced, a method to construct individual patient profiles is developed, and two dosing algorithms for oral and intestinal administration of levodopa are presented. Data from five different sources were used to develop the methods and evaluate the performance of the proposed algorithms.

    The dose automation algorithms can work both with clinical and objective ratings (through wearable devices), and their application was evaluated against dosing adjustments from movement disorders experts. Both dosing algorithms showed promise and their dosing suggestions were similar to those of the clinicians.

    The objective ratings algorithm had good test-retest reliability and its application during a clinical study was successful. Furthermore, the method of fitting individual patient models was robust and worked well with the objective ratings algorithm. Finally, a study was carried out that showed that about half the patients on levodopa treatment show reduced response during the afternoon hours, pointing to the need for more precise modelling of levodopa pharmacodynamics.

  • 118.
    Thomas, Ilias
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Optimizing levodopa dosing routines for Parkinson’s disease2017Licentiatavhandling, sammanläggning (Övrigt vetenskapligt)
    Abstract [en]

    This thesis in the field of microdata analysis aims to introduce dose optimizing algorithms for the pharmacological management of Parkinson’s disease (PD). PD is a neurodegenerative disease that mostly affects the motor functions of the patients and it is characterized as a movement disorder. The core symptoms of PD are: bradykinesia, postural instability, rigidity, and tremor. There is no cure for PD and the use of levodopa to manage the core symptoms is considered the gold standard. However, long term use of levodopa causes reduced medication efficacy, and side effects, such as dyskinesia, which can also be attributed to overmedication. When that happens precise individualized dosing schedules are required. The goal of this thesis is to examine if algorithmic methods can be used to find dosing schedules that treat PD symptoms and minimize manifestation of side effects. Data from three different sources were used for that purpose: data from a clinical study in Uppsala University hospital in 2015, patient admission chart data from Uppsala University hospital during 2011-2015, and data from a clinical study in Gothenburg University during 2016-2017. The data were used to develop the methods and evaluate the performance of the proposed algorithms.The first algorithm that was developed was a sensor-based method that derives objective measurements (ratings) of PD motor states. The construction of the sensor index was based on subjective ratings of patients’ motor functions made by three movement disorder experts. This sensor-based method was used when deriving algorithmic dosing schedules. Afterwards, a method that uses medication information and ratings of the patients’ motor states to fit individual patient models was developed. This method uses mathematical optimization to individualize specific parameters of dose-effects models for levodopa intake, through minimizing the distance between motor state ratings and dose-effect curves. Finally, two different dose optimization algorithms were developed and evaluated, that had as input the individual patient models. The first algorithm was specific to continuous infusion of levodopa treatment, where the patient’s state was set to a specific target value and the algorithm made dosing adjustments to keep that patients motor functions on that state. The second algorithm concerned oral administration of microtables of levodopa. The ambition with this algorithm was that the suggested doses would find the right balance between treating the core symptoms of PD and, at the same time, minimizing the side effects of long term levodopa use, mainly dyskinesia. Motor state ratings for this study were obtained through the sensor index. Both algorithms followed a principle of deriving a morning dose and a maintenance dose for the patients, with maintenance dose being an infusion rate for the first algorithm, and oral administration doses at specific time points for the second algorithm.The results showed that the sensor-based index had good test-retest reliability, sensitivity to levodopa treatment, and ability to make predictions in unseen parts of the dataset. The dosing algorithm for continuous infusion of levodopa had a good ability to suggest an optimal infusion rating for the patients, but consistently suggested lower morning dose than what the treating personnel prescribed. The dosing algorithm for oral administration of levodopa showed great agreement with the treating personnel’s prescriptions, both in terms of morning and maintenance dose. Moreover, when evaluating the oral medication algorithm, it was clear that the sensor index ratings could be used for building patient specific models.

  • 119.
    Thomas, Ilias
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Alam, Moudud
    Högskolan Dalarna, Akademin Industri och samhälle, Statistik.
    Bergquist, Filip
    Johansson, Dongni
    Memedi, Mevludin
    Nyholm, Dag
    Westin, Jerker
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    Sensor-based algorithmic dosing suggestions for oral administration of levodopa/carbidopa microtablets for Parkinson's disease: a first experience2019Ingår i: Journal of Neurology, ISSN 0340-5354, E-ISSN 1432-1459, Vol. 266, nr 3, s. 651-658Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    OBJECTIVE: Dosing schedules for oral levodopa in advanced stages of Parkinson's disease (PD) require careful tailoring to fit the needs of each patient. This study proposes a dosing algorithm for oral administration of levodopa and evaluates its integration into a sensor-based dosing system (SBDS).

    MATERIALS AND METHODS: In collaboration with two movement disorder experts a knowledge-driven, simulation based algorithm was designed and integrated into a SBDS. The SBDS uses data from wearable sensors to fit individual patient models, which are then used as input to the dosing algorithm. To access the feasibility of using the SBDS in clinical practice its performance was evaluated during a clinical experiment where dosing optimization of oral levodopa was explored. The supervising neurologist made dosing adjustments based on data from the Parkinson's KinetiGraph™ (PKG) that the patients wore for a week in a free living setting. The dosing suggestions of the SBDS were compared with the PKG-guided adjustments.

    RESULTS: The SBDS maintenance and morning dosing suggestions had a Pearson's correlation of 0.80 and 0.95 (with mean relative errors of 21% and 12.5%), to the PKG-guided dosing adjustments. Paired t test indicated no statistical differences between the algorithmic suggestions and the clinician's adjustments.

    CONCLUSION: This study shows that it is possible to use algorithmic sensor-based dosing adjustments to optimize treatment with oral medication for PD patients.

  • 120.
    Thomas, Ilias
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Alam, Moudud
    Högskolan Dalarna, Akademin Industri och samhälle, Statistik.
    Nyholm, Dag
    Senek, Marina
    Westin, Jerker
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    Individual dose-response models for levodopa infusion dose optimization2018Ingår i: International Journal of Medical Informatics, ISSN 1386-5056, E-ISSN 1872-8243, Vol. 112, s. 137-142Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Background and Objective

    To achieve optimal effect with continuous infusion treatment in Parkinson’s disease (PD), the individual doses (morning dose and continuous infusion rate) are titrated by trained medical personnel. This study describes an algorithmic method to derive optimized dosing suggestions for infusion treatment of PD, by fitting individual dose-response models. The feasibility of the proposed method was investigated using patient chart data.

    Methods

    Patient records were collected at Uppsala University hospital which provided dosing information and dose-response evaluations. Mathematical optimization was used to fit individual patient models using the records’ information, by minimizing an objective function. The individual models were passed to a dose optimization algorithm, which derived an optimized dosing suggestion for each patient model.

    Results

    Using data from a single day’s admission the algorithm showed great ability to fit appropriate individual patient models and derive optimized doses. The infusion rate dosing suggestions had 0.88 correlation and 10% absolute mean relative error compared to the optimal doses as determined by the hospital’s treating team. The morning dose suggestions were consistency lower that the optimal morning doses, which could be attributed to different dosing strategies and/or lack of on-off evaluations in the morning.

    Conclusion

    The proposed method showed promise and could be applied in clinical practice, to provide the hospital personnel with additional information when making dose adjustment decisions.

  • 121.
    Thomas, Ilias
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Bergquist, Filip
    Gothenburg University.
    Constantinescu, Radu
    Gothenburg University.
    Nyholm, Dag
    Dept. of Neuroscience, Neurology, Uppsala University.
    Senek, Marina
    Dept. of Neuroscience, Neurology, Uppsala University.
    Memedi, Mevludin
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik. Informatics, School of Business, Örebro University.
    Using measurements from wearable sensors for automatic scoring of Parkinson's disease motor states: Results from 7 patients2017Ingår i: Engineering in Medicine and Biology Society (EMBC), 2017 39th Annual International Conference of the IEEE, IEEE, 2017, s. 131-134Konferensbidrag (Refereegranskat)
    Abstract [en]

    The objective of this study was to investigate the validity of an objective gait measure for assessment of different motor states of advanced Parkinson's disease (PD) patients. Seven PD patients performed a gait task up to 15 times while wearing sensors on their upper and lower limbs. Each task was performed at specific points during a test day, following a single dose of levodopa-carbidopa. At the time of the tasks the patients were video recorded and three movement disorder experts rated their motor function on three clinical scales: a treatment response scale (TRS) that ranged from −3 (very bradykinetic) to 0 (ON) to +3 (very dyskinetic), a dyskinesia score that ranged from 0 (no dyskinesia) to 4 (extreme dyskinesia), and a bradykinesia score that ranged from 0 (no bradykinesia) to 4 (extreme bradykinesia). Raw accelerometer and gyroscope data of the sensors were processed and analyzed with time series analysis methods to extract features. The utilized features quantified separate limb movements as well as movement symmetries between the limbs. The features were processed with principal component analysis and the components were used as predictors for separate support vector machine (SVM) models for each of the three scales. The performance of each model was evaluated in a leave-one-patient out setting where the observations of a single patient were used as the testing set and the observations of the other 6 patients as the training set. Root mean square error (RMSE) and correlation coefficients for the predictions showed a good ability of the models to map the sensor data into the rating scales. There were strong correlations between the SVM models and the mean ratings of TRS (0.79; RMSE=0.70), bradykinesia score (0.79; RMSE=0.47), and bradykinesia score (0.78; RMSE=0.46). The results presented in this paper indicate that the use of wearable sensors when performing gait tasks can generate measurements that have a good correlation to subjective expert assessments.

  • 122.
    Thomas, Ilias
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Memedi, Mevludin
    Westin, Jerker
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    Nyholm, Dag
    The effect of continuous levodopa treatment during the afternoon hours2019Ingår i: Acta Neurologica Scandinavica, ISSN 0001-6314, E-ISSN 1600-0404, Vol. 139, nr 1, s. 70-75Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    OBJECTIVE: The aim of this retrospective study was to investigate if patients with PD, who are treated with levodopa-carbidopa intestinal gel (LCIG), clinically worsen during the afternoon hours and if so, to evaluate whether this occurs in all LCIG-treated patients or in a sub-group of patients.

    METHODS: Three published studies were identified and included in the analysis. All studies provided individual response data assessed on the treatment response scale (TRS) and patients were treated with continuous LCIG. Ninety-eight patients from the three studies fulfilled the criteria. T-tests were performed to find differences on the TRS values between the morning and the afternoon hours, linear mixed effect models were fitted on the afternoon hours' evaluations to find trends of wearing-off, and patients were classified into three TRS categories (meaningful increase in TRS, meaningful decrease in TRS, non -meaningful increase or decrease).

    RESULTS: In all three studies significant statistical differences were found between the morning TRS values and the afternoon TRS values (p-value <= 0.001 in all studies). The linear mixed effect models had significant negative coefficients for time in two studies, and 48 out of 98 patients (49%) showed a meaningful decrease of TRS during the afternoon hours.

    CONCLUSION: The results from all studies were consistent, both in the proportion of patients in the three groups and the value of TRS decrease in the afternoon hours. Based on these findings there seems to be a group of patients with predictable "off" behavior in the later parts of the day. This article is protected by copyright. All rights reserved.

  • 123.
    Thomas, Ilias
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Westin, Jerker
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    Alam, Moudud
    Högskolan Dalarna, Akademin Industri och samhälle, Statistik.
    Bergquist, F.
    Nyholm, D.
    Senek, M.
    Memedi, Mevludin
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    A treatment–response index from wearable sensors for quantifying Parkinson's disease motor states2018Ingår i: IEEE journal of biomedical and health informatics, ISSN 2168-2194, E-ISSN 2168-2208, Vol. 22, nr 5, s. 1341-1349Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    The goal of this study was to develop an algorithm that automatically quantifies motor states (off,on,dyskinesia) in Parkinson's disease (PD), based on accelerometry during a hand pronation-supination test. Clinician's ratings using the Treatment Response Scale (TRS), ranging from -3 (very Off) to 0 (On) to +3 (very dyskinetic), was used as target. For that purpose, 19 participants with advanced PD and 22 healthy persons were recruited in a single center open label clinical trial in Uppsala, Sweden. The trial consisted of single levodopa dose experiments for the people with PD (PwP), where participants were asked to perform standardized wrist rotation tests, using each hand, before and at pre-specified time points after the dose. The participants used wrist sensors containing a 3D accelerometer and gyroscope. Features to quantify the level, variation and asymmetry of the sensor signals, three-level Discrete Wavelet Transform features and approximate entropy measures were extracted from the sensors data. At the time of the tests, the PwP were video recorded. Three movement disorder specialists rated the participants’ state on the TRS scale. A Treatment Response Index from Sensors (TRIS) was constructed to quantify the motor states based on the wrist rotation tests. Different machine learning algorithms were evaluated to map the features derived from the sensor data to the ratings provided by the three specialists. Results from cross validation, both in 10-fold and a leave-one-individual out setting, showed good predictive power of a support vector machine model and high correlation to the TRS scale. Values at the end tails of the TRS scale were under and over predicted due to the lack of observations at those values but the model managed to accurately capture the dose - effect profiles of the patients. In addition, the TRIS had good test-retest reliability on the baseline levels of the PD participants (Intraclass correlation coefficient of 0.83) and reasonable sensitivity to levodopa treatment (0.33 for the TRIS). For a series of test occasions the proposed algorithms provided dose - effect time profiles for participants with PD, which could be useful during therapy individualization of people suffering from advanced PD

  • 124.
    Vasile, Manea
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Do administrative borders affect the long-term population dynamics?2015Självständigt arbete på avancerad nivå (masterexamen), 10 poäng / 15 hpStudentuppsats (Examensarbete)
    Abstract [en]

    How borders affect human activities like economic life has been a frequent research topic during the last decades. A widespread view is that local societies are affected by the proximity of their borders in one way or another. Within a specific country, research on border issues often has a center – periphery perspective, problematizing sub-national administrative decision-making as well as local and regional economic development. Besides international migration, it is not thoroughly investigated how demographical processes are affected by borders, although population change is of crucial importance for economic development. The aim in this study is therefore to analyze how local population changes in localities close to administrative borders are affected by local population change in the center. With this purpose, the causal time-space relationship is estimated. In this study, for a regional administrative unit (county), the population change in the smallest available administrative units (parishes) closely located to a county border are compared to the population change in those parishes located in the center in a county. This study builds on a long term time perspective and uses information on population totals in parishes in Sweden between 1810 and 2000. The long-term time perspective is needed because population change normally is a slow process. The findings show that there is a border effect and, on average, the population of Sweden tends to migrate from border locations to center locations within administrative borders. However, this process is region-dependent and is opposite in the southern counties, where the population migrates from the central towards the border locations, which most often are located at the coast.

  • 125.
    Wahab, Nor-Ul
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Evaluation of Supervised Machine LearningAlgorithms for Detecting Anomalies in Vehicle’s Off-Board Sensor Data2018Självständigt arbete på avancerad nivå (masterexamen), 20 poäng / 30 hpStudentuppsats (Examensarbete)
    Abstract [en]

    A diesel particulate filter (DPF) is designed to physically remove diesel particulate matter or soot from the exhaust gas of a diesel engine. Frequently replacing DPF is a waste of resource and waiting for full utilization is risky and very costly, so, what is the optimal time/milage to change DPF? Answering this question is very difficult without knowing when the DPF is changed in a vehicle.

    We are finding the answer with supervised machine learning algorithms for detecting anomalies in vehicles off-board sensor data (operational data of vehicles). Filter change is considered an anomaly because it is rare as compared to normal data.

    Non-sequential machine learning algorithms for anomaly detection like oneclass support vector machine (OC-SVM), k-nearest neighbor (K-NN), and random forest (RF) are applied for the first time on DPF dataset. The dataset is unbalanced, and accuracy is found misleading as a performance measure for the algorithms. Precision, recall, and F1-score are found good measure for the performance of the machine learning algorithms when the data is unbalanced. RF gave highest F1-score of 0.55 than K-NN (0.52) and OCSVM (0.51). It means that RF perform better than K-NN and OC-SVM but after further investigation it is concluded that the results are not satisfactory. However, a sequential approach should have been tried which could yield better result.

  • 126. Wei, Yixuan
    et al.
    Zhang, Xingxing
    Högskolan Dalarna, Akademin Industri och samhälle, Energiteknik.
    Shi, Yong
    Xia, Liang
    Pan, Song
    Wu, Jinshun
    Han, Mengjie
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Zhao, Xiaoyun
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    A review of data-driven approaches for prediction and classification of building energy consumption2018Ingår i: Renewable & sustainable energy reviews, ISSN 1364-0321, E-ISSN 1879-0690, Vol. 82, nr 1, s. 1027-1047Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    A recent surge of interest in building energy consumption has generated a tremendous amount of energy data, which boosts the data-driven algorithms for broad application throughout the building industry. This article reviews the prevailing data-driven approaches used in building energy analysis under different archetypes and granularities, including those methods for prediction (artificial neural networks, support vector machines, statistical regression, decision tree and genetic algorithm) and those methods for classification (K-mean clustering, self-organizing map and hierarchy clustering). The review results demonstrate that the data-driven approaches have well addressed a large variety of building energy related applications, such as load forecasting and prediction, energy pattern profiling, regional energy-consumption mapping, benchmarking for building stocks, global retrofit strategies and guideline making etc. Significantly, this review refines a few key tasks for modification of the data-driven approaches in the context of application to building energy analysis. The conclusions drawn in this review could facilitate future micro-scale changes of energy use for a particular building through the appropriate retrofit and the inclusion of renewable energy technologies. It also paves an avenue to explore potential in macro-scale energy-reduction with consideration of customer demands. All these will be useful to establish a better long-term strategy for urban sustainability.

  • 127.
    Yang, Bowen
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Ranking hotels for recommendation via generalized linear mixed model and Box-Cox model: A case of Stockholm rating data from booking.com2014Självständigt arbete på avancerad nivå (masterexamen), 20 poäng / 30 hpStudentuppsats (Examensarbete)
    Abstract [en]

    This thesis is aimed at recommending suitable hotels to the customers using the data collected from the website booking.com. In this thesis, data from Stockholm is chosen as an example, and statistical modeling is applied. We propose recommended hotels based on their rankings in terms of the scores of the hotels. The ranking score is derived by using eneralized linear mixed models. Box-Cox transformation is applied further to improve the previous analysis. Separate group analysis indicates that the ranks between different reviewer groups are significantly different. Model evaluation is executed via Cross-validation method by calculating the classification accuracies for all models. The best model is found based on theclassification accuracy, and we recommend the top 10, top 15 and top 20 hotels from the best model in this thesis.

  • 128. Zalakeviciute, R.
    et al.
    Rybarczyk, Yves
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys. Intelligent and Interactive Systems Lab (SI2 Lab) Universidad de Las Americas (UDLA), Ecuador.
    Granda-Albuja, M. G.
    Diaz Suarez, M. V.
    Alexandrino, K.
    Chemical characterization of urban PM10 in the Tropical Andes2019Ingår i: Atmospheric Pollution Research, ISSN 1309-1042, E-ISSN 1309-1042Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Complex inhalable particles have become one of the main causes to trigger health problems worldwide. While the level of concern depends on the chemical composition of these particles, some regions are poorly studied, particularly, the Andes. In this work, the chemical characterization of atmospheric PM10 filter samples, collected between January and October of 2017, was carried out for the first time in the world's highest capital, Quito, Ecuador. This study investigates PM10 relation with meteorological variables and criteria pollutants. Average PM10 concentrations ranged from 24.9 μg m−3 to 26.2 μg m−3, with some alarming peaks during the episodes of fires and New Year's celebration. The major elements at study sites were Ca, Na, S, Mg, P, K, Fe, Si and Al, while the major water-soluble ion was SO42−. Meteorology plays an important role at this complex terrain city. Factor analysis showed natural dust and soil resuspension as the main source of particulate matter. Moreover, two less urbanized sites showed evidence of industrial activities or airport emissions, while the central city site showed a very strong signal of traffic-related pollution. These results are compared with representative cities around the world. As is the case in developing countries, low-quality diesel fuel is recognized for emitting large amounts of heavy metals, resulting in higher levels of those tracers in traffic flow areas. This work demonstrates the problems facing a midsize city, such as the lack of stricter regulations and, thus compromised air quality. This may imply serious respiratory and cardiovascular health effects.

  • 129.
    Zhang, Fan
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    A hybrid structured deep neural network with Word2Vec for construction accident causes classification2019Ingår i: International Journal of Construction Management, ISSN 1562-3599Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    According to the latest fatal work injury rates reported by the Bureau of Labors Statistics, construction sites remain the most hazardous workplaces. In the construction sector, fatality investigation summary reports are available for past accidents and by investigating such reports, valuable insights can be gained. In this study, text mining algorithms are explored for automatic construction accident causes classification. To be more specific, Word2Vec skip-gram model is utilized to learn word embedding from a domain-specific corpus and a hybrid structured deep neural network is proposed by incorporating the learned word embedding for accident reports classification. Dataset from Occupational Safety and Health Administration (OSHA) is employed in the experiment to evaluate the performance of the proposed approach. Besides, five baseline models: support vector machine (SVM), linear regression (LR), K-nearest neighbor (KNN), decision tree (DT), Naive Bayes (NB) are employed to compare with the proposed approach. Experiment results show that the proposed model achieves the highest average weighted F1 score among all models considered in this study. The result also proves the effectiveness of applying Word2Vec skip-gram algorithm for semantic information augmentation. As a result, robustness of the model is improved when classifying cases of low support values.

  • 130.
    Zhang, Fan
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Fleyeh, Hasan
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    A Review of Single Artificial Neural Network Models for Electricity Spot Price Forecasting2019Ingår i: European Energy Market 2019, 2019Konferensbidrag (Refereegranskat)
    Abstract [en]

    Electricity price forecasting plays a crucial role in a liberalized electricity market. In terms of forecasting approaches, computational intelligence based models have been widely used with respect to electricity price forecasting and among all computation intelligence based models, artificial neural networks are most popular among researchers due to their flexibility and efficiency in handling complexity and non-linearity. However, a review of recent applications of neural networks for electricity price forecasting is not found in the literature. The motivation of this paper is to fill this research gap. In this study, existing approaches are analyzed and a summary of the strengths and weaknesses of each approach is presented. Besides, each neural network model is briefly summarized, followed by reviews of the corresponding studies of each neural network with respect to electricity forecasting from year 2010 onwards. Major contributions, datasets adopted as well as the corresponding experiment results are analyzed for each reviewed study. Apart from the review of existing studies, the advantages and disadvantages of each type of neural network model are discussed in details. Compared with neural networks based hybrid models, a single neural network model is easier to be implemented, less complex and more efficient. Scope of the review is the application of non-hybrid neural network models. It is found that most literature focuses on short term electricity price forecasting while medium and long term forecasting still remain relatively uncovered.

  • 131.
    Zhang, Fan
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Fleyeh, Hasan
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    A review on electricity price forecasting using neural network based models2018Rapport (Övrig (populärvetenskap, debatt, mm))
  • 132.
    Zhang, Fan
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Fleyeh, Hasan
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    Short Term Electricity Spot Price Forecasting Using CatBoost and Bidirectional Long Short Term Memory Neural Network2019Ingår i: 16th European Energy Market Conference (EEM 19), 2019Konferensbidrag (Refereegranskat)
    Abstract [en]

    Electricity price forecasting plays a crucial role in liberalized electricity markets. Generally speaking, short term electricity price forecasting is essential for electricity providers to adjust the schedule of production in order to balance consumers’ demands and electricity generation. Short term forecasting results are also utilized by market players to decide the timing of purchasing or selling to gain maximized profit. Among existing forecasting approaches, neural networks are regarded as the state of art method. However, deep neural networks are not studied comprehensively in this field, thus the motivation of this study is to fill this research gap. In this paper, a novel hybrid approach is proposed for short term electricity price forecasting. To be more specific, categorical boosting (Catboost) algorithm is used for feature selection and a bidirectional long short term memory neural network (BDLSTM) serves as the main forecasting engine. To evaluate the effectiveness of the proposed approach, two datasets from the Nord Pool market are employed in the experiment. Moreover, the performance of multi-layer perception (MLP) neural network, support vector regression (SVR) and ensemble tree models are evaluated and compared with the proposed model. Results show that the proposed approach outperforms the rest models in terms of mean absolute percentage error (MAPE).

  • 133.
    Zhang, Fan
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Fleyeh, Hasan
    Högskolan Dalarna, Akademin Industri och samhälle, Datateknik.
    Wang, X.
    Lu, M.
    Construction site accident analysis using text mining and natural language processing techniques2019Ingår i: Automation in Construction, ISSN 0926-5805, E-ISSN 1872-7891, Vol. 99, s. 238-248Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    Workplace safety is a major concern in many countries. Among various industries, construction sector is identified as the most hazardous work place. Construction accidents not only cause human sufferings but also result in huge financial loss. To prevent reoccurrence of similar accidents in the future and make scientific risk control plans, analysis of accidents is essential. In construction industry, fatality and catastrophe investigation summary reports are available for the past accidents. In this study, text mining and natural language process (NLP) techniques are applied to analyze the construction accident reports. To be more specific, five baseline models, support vector machine (SVM), linear regression (LR), K-nearest neighbor (KNN), decision tree (DT), Naive Bayes (NB) and an ensemble model are proposed to classify the causes of the accidents. Besides, Sequential Quadratic Programming (SQP) algorithm is utilized to optimize weight of each classifier involved in the ensemble model. Experiment results show that the optimized ensemble model outperforms rest models considered in this study in terms of average weighted F1 score. The result also shows that the proposed approach is more robust to cases of low support. Moreover, an unsupervised chunking approach is proposed to extract common objects which cause the accidents based on grammar rules identified in the reports. As harmful objects are one of the major factors leading to construction accidents, identifying such objects is extremely helpful to mitigate potential risks. Certain limitations of the proposed methods are discussed and suggestions and future improvements are provided.

  • 134.
    Zhang, Xingxing
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Energiteknik.
    Lovati, Marco
    Vigna, Ilaria
    Widén, Joakim
    Han, Mengjie
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Gál, Csilla V
    Högskolan Dalarna, Akademin Industri och samhälle, Energiteknik.
    Feng, Tao
    A review of urban energy systems at building cluster level incorporating renewable-energy-source (RES) envelope solutions2018Ingår i: Applied Energy, ISSN 0306-2619, E-ISSN 1872-9118, Vol. 230, s. 1034-1056Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    The emergence of renewable-energy-source (RES) envelope solutions, building retrofit requirements and advanced energy technologies brought about challenges to the existing paradigm of urban energy systems. It is envisioned that the building cluster approach—that can maximize the synergies of RES harvesting, building performance, and distributed energy management—will deliver the breakthrough to these challenges. Thus, this paper aims to critically review urban energy systems at the cluster level that incorporate building integrated RES solutions. We begin with defining cluster approach and the associated boundaries. Several factors influencing energy planning at cluster scale are identified, while the most important ones are discussed in detail. The closely reviewed factors include RES envelope solutions, solar energy potential, density of buildings, energy demand, integrated cluster-scale energy systems and energy hub. The examined categories of RES envelope solutions are (i) the solar power, (ii) the solar thermal and (iii) the energy-efficient ones, out of which solar energy is the most prevalent RES. As a result, methods assessing the solar energy potentials of building envelopes are reviewed in detail. Building density and the associated energy use are also identified as key factors since they affect the type and the energy harvesting potentials of RES envelopes. Modelling techniques for building energy demand at cluster level and their coupling with complex integrated energy systems or an energy hub are reviewed in a comprehensive way. In addition, the paper discusses control and operational methods as well as related optimization algorithms for the energy hub concept. Based on the findings of the review, we put forward a matrix of recommendations for cluster-level energy system simulations aiming to maximize the direct and indirect benefits of RES envelope solutions. By reviewing key factors and modelling approaches for characterizing RES-envelope-solutions-based urban energy systems at cluster level, this paper hopes to foster the transition towards more sustainable urban energy systems.

  • 135.
    Zhang, Xingxing
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Energiteknik.
    Wang, X.
    Zhao, Xiaoyun
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Solar System Design and Energy Performance Assessment Approaches2019Ingår i: Advanced Energy Efficiency Technologies for Solar Heating, Cooling and Power Generation, Springer, 2019, s. 417-451Kapitel i bok, del av antologi (Refereegranskat)
    Abstract [en]

    Recently, solar system has gained a rapid development in many countries because it is clean and sustainable. Many solar systems including the solar photovoltaic/loop-heat-pipe (PV/LHP), solar loop-heat-pipe (LHP), solar photovoltaic/micro-channel heat pipe (PV/MCHP) system, and solar thermal facade system (STF) have been designed for energy saving. To assess these systems’ performance, there are many approaches such as energy and exergy assessment which is used in this chapter to analyze their performance. Besides the system design, the authors set up dedicated experimental models in combination with computer models to test the systems’ performance. Furthermore, some systems are compared with the conventional system, and the performance of these solar systems is better than the conventional system. In addition, these solar systems are applied in many real buildings and their performance is examined, the results show that the solar systems have more potential to boost the building energy efficiency and create the possibility of solar development in buildings. © 2019, Springer Nature Switzerland AG.

  • 136.
    Zhang, Xingxing
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Energiteknik.
    Wei, Y.
    He, W.
    Qiu, Z.
    Zhao, Xiaoyun
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Solar Systems’ Economic and Environmental Performance Assessment2019Ingår i: Advanced Energy Efficiency Technologies for Solar Heating, Cooling and Power Generation / [ed] Xudong Zhao, Xiaoli Ma, Springer, 2019, s. 453-486Kapitel i bok, del av antologi (Refereegranskat)
    Abstract [en]

    The economic and environmental performance assessment of the solar system plays a critical role in building design, operation and retrofit. A dedicated economic model is necessary to assess the investment feasibility on a new technology, which allows investors to decide on a profitable investment, compare investment projects and know about the benefits of the best investment. An environmental model is adopted to predict carbon emission reduction in the solar system relative to the traditional heating and electronic systems. This chapter introduced three up-to-date solar system models and corresponding assessments related to their applications, including solar photovoltaic/loop heat pipe (PV/LHP) heat pump water heating system, loop heat pipe-based solar thermal facade (LHP-STF), heat pump water heating system as well as solar thermal facade (STF). The research results will be able to assist in decision-making in implementation of the proposed PV/T technology and analyses of the associated economic and environmental benefits, thus contributing to realization of regional and global targets on fossil fuel energy saving and environmental sustainability.

  • 137.
    Zhang, Xingxing
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Energiteknik.
    Wu, J.
    Pan, S.
    Han, Mengjie
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    An economic analysis of the solar photovoltaic/thermal (PV/T) technologies in Sweden: A case study2019Ingår i: IOP Conference Series: Materials Science and Engineering, 2019, Vol. 556, nr 1, artikel-id 012002Konferensbidrag (Refereegranskat)
  • 138.
    Zhang, Xingxing
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Energiteknik.
    Xiao, M.
    He, W.
    Qiu, Z.
    Zhao, Xiaoyun
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Heat Pump Technologies and Their Applications in Solar Systems2019Ingår i: Advanced Energy Efficiency Technologies for Solar Heating, Cooling and Power Generation / [ed] Xudong Zhao, Xiaoli Ma, Springer, 2019, s. 311-339Kapitel i bok, del av antologi (Refereegranskat)
    Abstract [en]

    As the well known that global energy demand is on a trend of continuous growth, reducing energy demand and making good use of renewable energy are thought to be the major routes toward low carbon and sustainable future, in particular for the building sector. Compared to traditional gas-fired heating systems, heat pumps have been proved to be an energy-efficient heating technology which can save fossil fuel energy and consequently reduce CO2 emission. However, the most outstanding challenges for the application of heat pumps lie in their high demand for electrical power, and the insufficient heat transfer between the heat source and the refrigerant. To overcome these difficulties, a solar-assisted heat pump has been proposed to tackle these challenges. A solar-assisted heat pump combines a heat pump with a solar collector, enabling the use of solar energy to provide space heating and hot water for buildings. This chapter introduces heat pump technologies and their applications in solar systems. Two types of solar-assisted heat pump, direct and indirect expansion, are illustrated in details. This work has provided the fundamental research and experience for developing a solar heat pump system and contributing to a significant fossil fuel saving and carbon reduction in the global extent.

  • 139.
    Zhao, Xiaoyun
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Government vs Market in Sustainable Residential Development?: Microdata analysis of car travel, CO2 emission and residence location2017Doktorsavhandling, sammanläggning (Övrigt vetenskapligt)
    Abstract [en]

    Increasing car usage and travel demands between residential locations and destinations in order to fulfill the various needs of residents is a primary cause of CO2 emissions. To win the battle against climate change, a better understanding of the question relating to which urban residential form may most effectively mitigate the CO2 emissions is the key pathway.

    This dissertation is concerned with the above problem and it mainly considers three objectives in providing insights on answering the question. The first objective is to comprehensively and microscopically understand intra-urban car travel behavior. The second objective is to estimate the induced CO2 emissions from daily intra-urban car travel and to ex-ante evaluate residential plans. The third objective is to assess whether the governmental sustainable residential development objective is aligned with the objectives of the estate market actors. To explore the research questions related to the objectives, a microdata analysis process (data collection, data assessment and transformation, data storage, data analysis and decision-making) is applied and is found essential in gaining access to key variables in exploring the answer of a preferable urban form. The dissertation offers many new solutions to various technical aspects through a microdata analysis process.

    The primary contribution of this dissertation is that it outlines an operational model that comprehensively integrates the investors’ investment strategy, the residents’ choice behavior, and the governmental sustainability objective in the interest of making an ex-ante assessment of residential plans. This ex-ante assessment provides decision-support in sustainable residential development at foremost local level.

    The first finding from the implementation of the model on the case study is that the market actors’ objectives are, in general, aligned with the government’s sustainable residential development objective. The second finding indicates that re-shaping the urban form into a compact city is preferable in mitigating CO2 emissions, in spite of the fact that the case city is of a polycentric urban form. These findings provide support for those advocating the compact city as the ideal for sustainable residential development, and also provide foresight on settling the answer to the preferred re-shaping of urban forms in climate change.

  • 140.
    Zhao, Xiaoyun
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Road network and GPS tracking with data processing and quality assessment2015Licentiatavhandling, sammanläggning (Övrigt vetenskapligt)
    Abstract [en]

    GPS technology has been embedded into portable, low-cost electronic devices nowadays to track the movements of mobile objects. This implication has greatly impacted the transportation field by creating a novel and rich source of traffic data on the road network. Although the promise offered by GPS devices to overcome problems like underreporting, respondent fatigue, inaccuracies and other human errors in data collection is significant; the technology is still relatively new that it raises many issues for potential users. These issues tend to revolve around the following areas: reliability, data processing and the related application.

    This thesis aims to study the GPS tracking form the methodological, technical and practical aspects. It first evaluates the reliability of GPS based traffic data based on data from an experiment containing three different traffic modes (car, bike and bus) traveling along the road network. It then outline the general procedure for processing GPS tracking data and discuss related issues that are uncovered by using real-world GPS tracking data of 316 cars. Thirdly, it investigates the influence of road network density in finding optimal location for enhancing travel efficiency and decreasing travel cost.

    The results show that the geographical positioning is reliable. Velocity is slightly underestimated, whereas altitude measurements are unreliable.Post processing techniques with auxiliary information is found necessary and important when solving the inaccuracy of GPS data. The densities of the road network influence the finding of optimal locations. The influence will stabilize at a certain level and do not deteriorate when the node density is higher.

  • 141.
    Zhao, Xiaoyun
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Carling, Kenneth
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Håkansson, Johan
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    An evaluation of the reliability of GPS-based transportation data2017Ingår i: Proceedings of IAC in Vienna 2017, 2017, s. 323-334, artikel-id IAC201711035Konferensbidrag (Refereegranskat)
    Abstract [en]

    GPS-based data are becoming a cornerstone for real-time transportation applications. Tracking data of vehicles from GPS receivers are however susceptible to measurement errors. The assessment of the reliability of data from GPS receiver is a neglected issue, especially in a real road network setting and in the phase after data transfer but before information identification. An evaluation method is outlined and carried out by conducting a randomized experiment. We assess the reliability of GPS-based transportation data on geographical position, speed, and elevation from three varied receivers GlobalSat BT-338X, Magellan SporTrak Pro and smart phone for three transportation modes: bicycle, car, and bus. The positional error ranging from 0158 meters, and 74% to 100% with an error within 5 meters depending on the transportation mode and route, there is also a non-negligible risk for aberrant positioning. Speed is slightly underestimated or overestimated with errors around 5km/h except for SporTrak Pro which had an error of -10 km/h. Elevation measurements are unreliable with errors bigger than 100 meters.

  • 142.
    Zhao, Xiaoyun
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Statistik.
    Carling, Kenneth
    Högskolan Dalarna, Akademin Industri och samhälle, Statistik.
    Håkansson, Johan
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    An Evaluation of the Reliability of GPS-Based Transportation Data2014Rapport (Övrigt vetenskapligt)
    Abstract [en]

    GPS-based data are becoming a cornerstone for real-time transportation applications. Tracking data of vehicles from GPS receivers are however susceptible to measurement errors. The assessment of the reliability of data from GPS receiver is a neglected issue, especially in a real road network setting and in the phase after data transfer but before information identification. An evaluation method is outlined and carried out by conducting a randomized experiment. We assess the reliability of GPS-based transportation data on geographical position, speed, and elevation from three varied receivers GlobalSat BT-338X, Magellan SporTrak Pro and smart phone for three transportation modes: bicycle, car, and bus. The positional error ranging from 0158 meters, and 74% to 100% with an error within 5 meters depending on the transportation mode and route, there is also a non-negligible risk for aberrant positioning. Speed is slightly underestimated or overestimated with errors around 5km/h except for SporTrak Pro which had an error

  • 143.
    Zhao, Xiaoyun
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Carling, Kenneth
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Håkansson, Johan
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    On assessing governmental sustainable residential planning and its alignment with residents’ and estate investors’ objectives2017Rapport (Övrigt vetenskapligt)
    Abstract [en]

    There are three key actors in forming the sustainable spatial distribution of residency in an area, (local) government, the estate investor and the resident, each with its own objective. Most urban planning studies have mainly focused on the ex-post evaluation of residential development by considering the objective of each actor separately. This paper outlines a conceptual model where the three key actors and their unique objectives are integrated with the aim of providing an ex-ante evaluation of residential development for government to make policies operational on a micro level. The methodology is implemented on a Swedish city, where sustainable residential development is in high need due to the influx of immigrants. The case study demonstrates that the model can integrate the macro and micro actors well. The model can provide noteworthy insights for the government on where the objectives of sustainability, livability and profit can be met. A sensitivity check of the parameter settings shows that the implementation of the model is robust for replication in other cities.

  • 144.
    Zhao, Xiaoyun
    et al.
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Carling, Kenneth
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Håkansson, Johan
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    Residential planning, driver mobility and CO2 emission: a microscopic look at Borlänge in Sweden2017Ingår i: European Planning Studies, ISSN 0965-4313, E-ISSN 1469-5944, Vol. 25, nr 9, s. 1597-1614Artikel i tidskrift (Refereegranskat)
    Abstract [en]

    In a city there are hotspots that attract citizens, and most of the transportation arises when citizens move between their residence and primary destinations (i.e. hotspots). However, an ex ante evaluation of energy-efficient mobility and urban residential planning has seldom been conducted. Therefore, this paper proposes an ex ante evaluation method to quantify the impacts, in terms of CO2 emissions induced by intra-urban car mobility, of residential plans for various urban areas. The method is illustrated in a case study of a Swedish midsize city, which is presently preoccupied with urban planning of new residential areas in response to substantial population growth due to immigration. In general, CO2 emissions increase from the continued urban core area (CUCA), to the sub-polycentric area (SPA), to the edge urbanization area (EUA), where CO2 emission of EUA is twice that of the CUCA. The average travel distances also increase in the same pattern, though the relative increase is more than four times. Apartment buildings could be more effective in meeting residential needs and mitigating CO2 emissions than dispersed single-family houses. 

  • 145.
    Zou, Luyi
    Högskolan Dalarna, Akademin Industri och samhälle, Mikrodataanalys.
    An Investigation of the LDA based Topic Model Approachfor Data Mining Twitter Social Network2015Självständigt arbete på avancerad nivå (masterexamen), 10 poäng / 15 hpStudentuppsats (Examensarbete)
    Abstract [en]

    In recent years, Twitter has become a highly popular form of social media.Twitter provides a platform for users to post short messages for followers to read inan on-or off-line fashion. Twitter is used in a variety of ways, from posting aboutpersonal daily life, to keeping up to date with current events.This thesis aims to find a reliable pipeline to analyse and visualize hottest topics(or trends) that people are talking about on Twitter during a period of time. Topicmodel is used to cluster Twitter messages and identify topic words, then topic wordscombined with the tweets’ influences are graphically represented by visualizationsoftware to reflect the trend under the topic. However, two limitations of Twittermessages prevent normal topic model tools from being applied their full potentials:Twitter messages are short and and colloquial. Twitter message provides little usefulinformation for the topic model to work properly. Thus, we proposed an poolingschema to enhance the performance of a topic model on Twitter data. Meanwhile, toidentify a reliable pipeline to do the task, we compared different methodologiesduring the process. We compared performance with and without pooling schema inthe data sampling step, performance with and without TF*IDF in the data processingstep; and finally compare performance of Latent Dirichlet allocation (LDA) withCorrelated Topic Models (CTM) to identify a topic. The results show thatLDA-TF*IDF with pooling schema is the most accurate model to identify Twittertrend.

123 101 - 145 av 145
RefereraExporteraLänk till träfflistan
Permanent länk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • chicago-author-date
  • chicago-note-bibliography
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf