{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/88934"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/88934","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Maintaining privacy during continuous motion sensing","abstract":"Mobile devices contain sensors which allow continuous recording of a user's motion allowing the development of activity, fitness and health applications. With varied applications, the motion sensors present new privacy problems which require protection. This dissertation builds on previous work with activity and fitness machine learning techniques demonstrating the ability to predict medical values from motion data using smartphones. We conduct two clinical trials collecting a data set of eighty-eight patients and forty-five hours of monitoring to analyze the privacy implications of releasing motion data. We extract a comprehensive set of statistical features from all available smartphone sensors and evaluate feature selection techniques and machine learning models. We find we can predict user identity, phone identity, speed, FEV1/FVC, and activity from the motion signal. Designing a privacy protection mechanism for motion data requires a precise understanding of how the signal predicts the sensitive information. We develop algorithms to conduct private feature selection which identifies features useful for prediction. We find that simply blocking all private features significantly reduces the usefulness of the signal for other predictions. We develop a sensitivity estimation framework to calibrate the noise for each private feature requiring an order of magnitude less noise than differential privacy sensitivity. We find adding noise to private features calibrated using the sensitivity estimate is effective at reducing the prediction of five tested target predictions. Our methods hide both user and phone identification while allowing other prediction but cannot hide activity, FEV1/FVC and speed without significantly lowering the accuracy of other predictions. Our methods are still effective when the attacker has prior knowledge of the noise distribution. The methods presented in this dissertation demonstrate the need for privacy in motion data and provide a framework for protecting sensitive user information in motion readings.","abstract_html":"Mobile devices contain sensors which allow continuous recording of a user's motion allowing the development of activity, fitness and health applications. With varied applications, the motion sensors present new privacy problems which require protection. This dissertation builds on previous work with activity and fitness machine learning techniques demonstrating the ability to predict medical values from motion data using smartphones. We conduct two clinical trials collecting a data set of eighty-eight patients and forty-five hours of monitoring to analyze the privacy implications of releasing motion data. We extract a comprehensive set of statistical features from all available smartphone sensors and evaluate feature selection techniques and machine learning models. We find we can predict user identity, phone identity, speed, FEV1/FVC, and activity from the motion signal. Designing a privacy protection mechanism for motion data requires a precise understanding of how the signal predicts the sensitive information. We develop algorithms to conduct private feature selection which identifies features useful for prediction. We find that simply blocking all private features significantly reduces the usefulness of the signal for other predictions. We develop a sensitivity estimation framework to calibrate the noise for each private feature requiring an order of magnitude less noise than differential privacy sensitivity. We find adding noise to private features calibrated using the sensitivity estimate is effective at reducing the prediction of five tested target predictions. Our methods hide both user and phone identification while allowing other prediction but cannot hide activity, FEV1/FVC and speed without significantly lowering the accuracy of other predictions. Our methods are still effective when the attacker has prior knowledge of the noise distribution. The methods presented in this dissertation demonstrate the need for privacy in motion data and provide a framework for protecting sensitive user information in motion readings.","abstract_has_math":false,"creators":["Juen, Joshua Paul Joseph"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Electrical & Computer Engineering","degree_department":null,"school":null,"contributors":["Borisov, Nikita","Schatz, Bruce","Caesar, Matthew","Choudhury, Romit"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2016,"date_issued":"2016-03-02T19:32:53Z","date_published":"2016-03-02T19:32:53Z","updated_at":"2026-07-22T22:26:32Z","subjects":["Computer Security","Privacy","Machine Learning","Big Data","Phone Sensors","Smartphones","Differential Privacy"],"languages":["en"],"rights":["2015 Joshua Paul-Joseph Juen"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/88934","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Borisov, Nikita","Schatz, Bruce","Caesar, Matthew","Choudhury, Romit"]},{"key":"dc:creator","label":"Author","values":["Juen, Joshua Paul Joseph"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2016-03-02T19:32:53Z","2015-08-18","2015-12"]},{"key":"dc:type","label":"Dc Type","values":["text"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical & Computer Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph.D."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Computer Security","Privacy","Machine Learning","Big Data","Phone Sensors","Smartphones","Differential Privacy"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["2015 Joshua Paul-Joseph Juen"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/88934"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Mobile devices contain sensors which allow continuous recording of a user's motion allowing the development of activity, fitness and health applications. With varied applications, the motion sensors present new privacy problems which require protection. This dissertation builds on previous work with activity and fitness machine learning techniques demonstrating the ability to predict medical values from motion data using smartphones. We conduct two clinical trials collecting a data set of eighty-eight patients and forty-five hours of monitoring to analyze the privacy implications of releasing motion data. We extract a comprehensive set of statistical features from all available smartphone sensors and evaluate feature selection techniques and machine learning models. We find we can predict user identity, phone identity, speed, FEV1/FVC, and activity from the motion signal. Designing a privacy protection mechanism for motion data requires a precise understanding of how the signal predicts the sensitive information. We develop algorithms to conduct private feature selection which identifies features useful for prediction. We find that simply blocking all private features significantly reduces the usefulness of the signal for other predictions. We develop a sensitivity estimation framework to calibrate the noise for each private feature requiring an order of magnitude less noise than differential privacy sensitivity. We find adding noise to private features calibrated using the sensitivity estimate is effective at reducing the prediction of five tested target predictions. Our methods hide both user and phone identification while allowing other prediction but cannot hide activity, FEV1/FVC and speed without significantly lowering the accuracy of other predictions. Our methods are still effective when the attacker has prior knowledge of the noise distribution. The methods presented in this dissertation demonstrate the need for privacy in motion data and provide a framework for protecting sensitive user information in motion readings.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2016-03-02 without embargo terms","The student, Joshua Juen, accepted the attached license on 2015-08-17 at 10:10.","The student, Joshua Juen, submitted this Dissertation for approval on 2015-08-17 at 10:22.","This Dissertation was approved for publication on 2015-08-18 at 14:15.","DSpace SAF Submission Ingestion Package generated from Vireo submission #8661 on 2016-03-02 at 12:48:46","Made available in DSpace on 2016-03-02T19:32:53Z (GMT). No. of bitstreams: 3 JUEN-DISSERTATION-2015.pdf: 5678615 bytes, checksum: 36737b04b7fa1319cf192ce100e7804f (MD5) Juen_Joshua.zip: 12268423 bytes, checksum: 4ac3528eeeaa136297efb3548770a6a4 (MD5) LICENSE.txt: 4208 bytes, checksum: cabf14f841a47559243083d3743e1969 (MD5) Previous issue date: 2015-08-18"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Maintaining privacy during continuous motion sensing"]}]}],"canonical_facts":{"dc:contributor":["Borisov, Nikita","Schatz, Bruce","Caesar, Matthew","Choudhury, Romit"],"dc:creator":["Juen, Joshua Paul Joseph"],"dc:date":["2016-03-02T19:32:53Z","2015-08-18","2015-12"],"dc:description":["Mobile devices contain sensors which allow continuous recording of a user's motion allowing the development of activity, fitness and health applications. With varied applications, the motion sensors present new privacy problems which require protection. This dissertation builds on previous work with activity and fitness machine learning techniques demonstrating the ability to predict medical values from motion data using smartphones. We conduct two clinical trials collecting a data set of eighty-eight patients and forty-five hours of monitoring to analyze the privacy implications of releasing motion data. We extract a comprehensive set of statistical features from all available smartphone sensors and evaluate feature selection techniques and machine learning models. We find we can predict user identity, phone identity, speed, FEV1/FVC, and activity from the motion signal. Designing a privacy protection mechanism for motion data requires a precise understanding of how the signal predicts the sensitive information. We develop algorithms to conduct private feature selection which identifies features useful for prediction. We find that simply blocking all private features significantly reduces the usefulness of the signal for other predictions. We develop a sensitivity estimation framework to calibrate the noise for each private feature requiring an order of magnitude less noise than differential privacy sensitivity. We find adding noise to private features calibrated using the sensitivity estimate is effective at reducing the prediction of five tested target predictions. Our methods hide both user and phone identification while allowing other prediction but cannot hide activity, FEV1/FVC and speed without significantly lowering the accuracy of other predictions. Our methods are still effective when the attacker has prior knowledge of the noise distribution. The methods presented in this dissertation demonstrate the need for privacy in motion data and provide a framework for protecting sensitive user information in motion readings.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2016-03-02 without embargo terms","The student, Joshua Juen, accepted the attached license on 2015-08-17 at 10:10.","The student, Joshua Juen, submitted this Dissertation for approval on 2015-08-17 at 10:22.","This Dissertation was approved for publication on 2015-08-18 at 14:15.","DSpace SAF Submission Ingestion Package generated from Vireo submission #8661 on 2016-03-02 at 12:48:46","Made available in DSpace on 2016-03-02T19:32:53Z (GMT). No. of bitstreams: 3 JUEN-DISSERTATION-2015.pdf: 5678615 bytes, checksum: 36737b04b7fa1319cf192ce100e7804f (MD5) Juen_Joshua.zip: 12268423 bytes, checksum: 4ac3528eeeaa136297efb3548770a6a4 (MD5) LICENSE.txt: 4208 bytes, checksum: cabf14f841a47559243083d3743e1969 (MD5) Previous issue date: 2015-08-18"],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/88934"],"dc:language":["en"],"dc:rights":["2015 Joshua Paul-Joseph Juen"],"dc:subject":["Computer Security","Privacy","Machine Learning","Big Data","Phone Sensors","Smartphones","Differential Privacy"],"dc:title":["Maintaining privacy during continuous motion sensing"],"dc:type":["text"],"thesis:degree_discipline":["Electrical & Computer Engineering"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Ph.D."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:26:32Z"}