Semi-supervised segmentation of accelerometer time series for transport mode classification. Leodolter, M., Widhalm, P., Plant, C., & Brandle, N. In 2017 5th IEEE International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS), pages 663–668, June, 2017. doi abstract bibtex Collecting ground truth data with smart phone applications is as difficult as important for training classification models predicting transport modes of people. Errors of respondent input with respect to trip length and transport mode segmenting introduce a systematic bias in the classification model. We propose a semi-supervised framework adjusting user-given input to process user-collected accelerometer time series data. Our contributions are (1) an evaluation of the impact of segmentation bias, (2) a novel algorithm to find more homogeneous segments and (3) a robust incrementally trained classifier model based on clustering employing Dynamic Time Warping as similarity measure. We apply the proposed method on synthetic and real-world accelerometer trip data of 800 labeled trips consisting of 2000 user-given segments and 400 hours travel time and test it against a baseline classifier relying completely on user-feedback. The results prove that our method learns clusters revised from noise and increases the classifier's accuracy for real-world and synthetic data by up to 17%.
@inproceedings{leodolter_semi-supervised_2017,
title = {Semi-supervised segmentation of accelerometer time series for transport mode classification},
doi = {10.1109/MTITS.2017.8005596},
abstract = {Collecting ground truth data with smart phone applications is as difficult as important for training classification models predicting transport modes of people. Errors of respondent input with respect to trip length and transport mode segmenting introduce a systematic bias in the classification model. We propose a semi-supervised framework adjusting user-given input to process user-collected accelerometer time series data. Our contributions are (1) an evaluation of the impact of segmentation bias, (2) a novel algorithm to find more homogeneous segments and (3) a robust incrementally trained classifier model based on clustering employing Dynamic Time Warping as similarity measure. We apply the proposed method on synthetic and real-world accelerometer trip data of 800 labeled trips consisting of 2000 user-given segments and 400 hours travel time and test it against a baseline classifier relying completely on user-feedback. The results prove that our method learns clusters revised from noise and increases the classifier's accuracy for real-world and synthetic data by up to 17\%.},
booktitle = {2017 5th {IEEE} {International} {Conference} on {Models} and {Technologies} for {Intelligent} {Transportation} {Systems} ({MT}-{ITS})},
author = {Leodolter, M. and Widhalm, P. and Plant, C. and Brandle, N.},
month = jun,
year = {2017},
keywords = {Accelerometer, Accelerometers, Clustering, Clustering algorithms, Dynamic Time Warping, Feature extraction, Heuristic algorithms, Legged locomotion, Robustness, Segmentation, Time series analysis, Transport Mode Detection, accelerometers, dynamic time warping, ground truth data, homogeneous segments, labeled trips, learning (artificial intelligence), pattern classification, pattern clustering, real-world accelerometer trip data, robust incrementally trained classifier model, segmentation bias, semisupervised accelerometer time series segmentation, similarity measure, smart phone, smart phones, synthetic accelerometer trip data, time series, traffic information systems, transport mode classification, trip length, user-given segments},
pages = {663--668},
}
Downloads: 0
{"_id":"wwSZ8DeZ4tjxSzBsW","bibbaseid":"leodolter-widhalm-plant-brandle-semisupervisedsegmentationofaccelerometertimeseriesfortransportmodeclassification-2017","author_short":["Leodolter, M.","Widhalm, P.","Plant, C.","Brandle, N."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Semi-supervised segmentation of accelerometer time series for transport mode classification","doi":"10.1109/MTITS.2017.8005596","abstract":"Collecting ground truth data with smart phone applications is as difficult as important for training classification models predicting transport modes of people. Errors of respondent input with respect to trip length and transport mode segmenting introduce a systematic bias in the classification model. We propose a semi-supervised framework adjusting user-given input to process user-collected accelerometer time series data. Our contributions are (1) an evaluation of the impact of segmentation bias, (2) a novel algorithm to find more homogeneous segments and (3) a robust incrementally trained classifier model based on clustering employing Dynamic Time Warping as similarity measure. We apply the proposed method on synthetic and real-world accelerometer trip data of 800 labeled trips consisting of 2000 user-given segments and 400 hours travel time and test it against a baseline classifier relying completely on user-feedback. The results prove that our method learns clusters revised from noise and increases the classifier's accuracy for real-world and synthetic data by up to 17%.","booktitle":"2017 5th IEEE International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS)","author":[{"propositions":[],"lastnames":["Leodolter"],"firstnames":["M."],"suffixes":[]},{"propositions":[],"lastnames":["Widhalm"],"firstnames":["P."],"suffixes":[]},{"propositions":[],"lastnames":["Plant"],"firstnames":["C."],"suffixes":[]},{"propositions":[],"lastnames":["Brandle"],"firstnames":["N."],"suffixes":[]}],"month":"June","year":"2017","keywords":"Accelerometer, Accelerometers, Clustering, Clustering algorithms, Dynamic Time Warping, Feature extraction, Heuristic algorithms, Legged locomotion, Robustness, Segmentation, Time series analysis, Transport Mode Detection, accelerometers, dynamic time warping, ground truth data, homogeneous segments, labeled trips, learning (artificial intelligence), pattern classification, pattern clustering, real-world accelerometer trip data, robust incrementally trained classifier model, segmentation bias, semisupervised accelerometer time series segmentation, similarity measure, smart phone, smart phones, synthetic accelerometer trip data, time series, traffic information systems, transport mode classification, trip length, user-given segments","pages":"663–668","bibtex":"@inproceedings{leodolter_semi-supervised_2017,\n\ttitle = {Semi-supervised segmentation of accelerometer time series for transport mode classification},\n\tdoi = {10.1109/MTITS.2017.8005596},\n\tabstract = {Collecting ground truth data with smart phone applications is as difficult as important for training classification models predicting transport modes of people. Errors of respondent input with respect to trip length and transport mode segmenting introduce a systematic bias in the classification model. We propose a semi-supervised framework adjusting user-given input to process user-collected accelerometer time series data. Our contributions are (1) an evaluation of the impact of segmentation bias, (2) a novel algorithm to find more homogeneous segments and (3) a robust incrementally trained classifier model based on clustering employing Dynamic Time Warping as similarity measure. We apply the proposed method on synthetic and real-world accelerometer trip data of 800 labeled trips consisting of 2000 user-given segments and 400 hours travel time and test it against a baseline classifier relying completely on user-feedback. The results prove that our method learns clusters revised from noise and increases the classifier's accuracy for real-world and synthetic data by up to 17\\%.},\n\tbooktitle = {2017 5th {IEEE} {International} {Conference} on {Models} and {Technologies} for {Intelligent} {Transportation} {Systems} ({MT}-{ITS})},\n\tauthor = {Leodolter, M. and Widhalm, P. and Plant, C. and Brandle, N.},\n\tmonth = jun,\n\tyear = {2017},\n\tkeywords = {Accelerometer, Accelerometers, Clustering, Clustering algorithms, Dynamic Time Warping, Feature extraction, Heuristic algorithms, Legged locomotion, Robustness, Segmentation, Time series analysis, Transport Mode Detection, accelerometers, dynamic time warping, ground truth data, homogeneous segments, labeled trips, learning (artificial intelligence), pattern classification, pattern clustering, real-world accelerometer trip data, robust incrementally trained classifier model, segmentation bias, semisupervised accelerometer time series segmentation, similarity measure, smart phone, smart phones, synthetic accelerometer trip data, time series, traffic information systems, transport mode classification, trip length, user-given segments},\n\tpages = {663--668},\n}\n\n\n\n","author_short":["Leodolter, M.","Widhalm, P.","Plant, C.","Brandle, N."],"key":"leodolter_semi-supervised_2017","id":"leodolter_semi-supervised_2017","bibbaseid":"leodolter-widhalm-plant-brandle-semisupervisedsegmentationofaccelerometertimeseriesfortransportmodeclassification-2017","role":"author","urls":{},"keyword":["Accelerometer","Accelerometers","Clustering","Clustering algorithms","Dynamic Time Warping","Feature extraction","Heuristic algorithms","Legged locomotion","Robustness","Segmentation","Time series analysis","Transport Mode Detection","accelerometers","dynamic time warping","ground truth data","homogeneous segments","labeled trips","learning (artificial intelligence)","pattern classification","pattern clustering","real-world accelerometer trip data","robust incrementally trained classifier model","segmentation bias","semisupervised accelerometer time series segmentation","similarity measure","smart phone","smart phones","synthetic accelerometer trip data","time series","traffic information systems","transport mode classification","trip length","user-given segments"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/mh_lenguyen","dataSources":["iwKepCrWBps7ojhDx"],"keywords":["accelerometer","accelerometers","clustering","clustering algorithms","dynamic time warping","feature extraction","heuristic algorithms","legged locomotion","robustness","segmentation","time series analysis","transport mode detection","accelerometers","dynamic time warping","ground truth data","homogeneous segments","labeled trips","learning (artificial intelligence)","pattern classification","pattern clustering","real-world accelerometer trip data","robust incrementally trained classifier model","segmentation bias","semisupervised accelerometer time series segmentation","similarity measure","smart phone","smart phones","synthetic accelerometer trip data","time series","traffic information systems","transport mode classification","trip length","user-given segments"],"search_terms":["semi","supervised","segmentation","accelerometer","time","series","transport","mode","classification","leodolter","widhalm","plant","brandle"],"title":"Semi-supervised segmentation of accelerometer time series for transport mode classification","year":2017}