{"id":{"repo_id":"trento","oai_identifier":"oai:iris.unitn.it:11572/407669"},"canonical_url":"https://search.dev.ndltd.org/etd/trento/oai:iris.unitn.it:11572/407669","repository":{"repo_id":"trento","name":"Università degli Studi di Trento","base_url":"https://iris.unitn.it/oai/request"},"display":{"title":"Human-Machine Alignment for Context Recognition in the Wild","abstract":"The premise for AI systems like personal assistants to provide guidance and suggestions to an end-user is to understand, at any moment in time, the personal context that the user is in. The context – where the user is, what she is doing and with whom – allows the machine to represent the world in user’s terms. The context must be inferred from a stream of sensor readings generated by smart wearables such as smartphones and smartwatches, and the labels are acquired from the user directly. To perform robust context prediction in this real-world scenario, the machine must handle the egocentric nature of the context, adapt to the changing world and user, and maintain a bidirectional interaction with the user to ensure the user-machine alignment of world representations. To this end, the machine must learn incrementally on the input stream of sensor readings and user supervision. In this work, we: (i) introduce interactive classification in the wild and present knowledge drift (KD), a special form of concept drift, occurring due to world and user changes; (ii) develop simple and robust ML methods to tackle these scenarios; (iii) showcase the advantages of each of these methods in empirical evaluations on controlled synthetic and real-world data sets; (iv) design a flexible and modular architecture that combines the methods above to support context recognition in the wild; (v) present an evaluation with real users in a concrete social science use case.","abstract_html":"The premise for AI systems like personal assistants to provide guidance and suggestions to an end-user is to understand, at any moment in time, the personal context that the user is in. The context – where the user is, what she is doing and with whom – allows the machine to represent the world in user’s terms. The context must be inferred from a stream of sensor readings generated by smart wearables such as smartphones and smartwatches, and the labels are acquired from the user directly. To perform robust context prediction in this real-world scenario, the machine must handle the egocentric nature of the context, adapt to the changing world and user, and maintain a bidirectional interaction with the user to ensure the user-machine alignment of world representations. To this end, the machine must learn incrementally on the input stream of sensor readings and user supervision. In this work, we: (i) introduce interactive classification in the wild and present knowledge drift (KD), a special form of concept drift, occurring due to world and user changes; (ii) develop simple and robust ML methods to tackle these scenarios; (iii) showcase the advantages of each of these methods in empirical evaluations on controlled synthetic and real-world data sets; (iv) design a flexible and modular architecture that combines the methods above to support context recognition in the wild; (v) present an evaluation with real users in a concrete social science use case.","abstract_has_math":false,"creators":["Bontempelli, Andrea"],"institution":"Università degli studi di Trento","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Giunchiglia, Fausto","Passerini, Andrea"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024-04-30","date_published":"2024-04-30","updated_at":"2026-07-24T05:04:24Z","subjects":["interactive machine learning, incremental learning, data streams human-in-the-loop, concept drift, context recognition"],"languages":["eng"],"rights":["info:eu-repo/semantics/openAccess","license:Creative commons","license uri:http://creativecommons.org/licenses/by-sa/4.0/"],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["http://dx.doi.org/10.15168/11572_407669","10.15168/11572_407669"],"render_values":[{"text":"http://dx.doi.org/10.15168/11572_407669","href":"http://dx.doi.org/10.15168/11572_407669","code":true},{"text":"10.15168/11572_407669","href":"https://doi.org/10.15168/11572_407669","code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/11572/407669","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Bontempelli, Andrea","Giunchiglia, Fausto","Passerini, Andrea"]},{"key":"dc:creator","label":"Author","values":["Bontempelli, Andrea"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2024-04-30"]},{"key":"dc:publisher","label":"Institution","values":["Università degli studi di Trento","place:TRENTO"]},{"key":"dc:relation","label":"Dc Relation","values":["firstpage:1","lastpage:139","numberofpages:139"]},{"key":"dc:type","label":"Dc Type","values":["info:eu-repo/semantics/doctoralThesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["interactive machine learning, incremental learning, data streams human-in-the-loop, concept drift, context recognition"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["info:eu-repo/semantics/openAccess","license:Creative commons","license uri:http://creativecommons.org/licenses/by-sa/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://hdl.handle.net/11572/407669","http://dx.doi.org/10.15168/11572_407669","10.15168/11572_407669"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["The premise for AI systems like personal assistants to provide guidance and suggestions to an end-user is to understand, at any moment in time, the personal context that the user is in. The context – where the user is, what she is doing and with whom – allows the machine to represent the world in user’s terms. The context must be inferred from a stream of sensor readings generated by smart wearables such as smartphones and smartwatches, and the labels are acquired from the user directly. To perform robust context prediction in this real-world scenario, the machine must handle the egocentric nature of the context, adapt to the changing world and user, and maintain a bidirectional interaction with the user to ensure the user-machine alignment of world representations. To this end, the machine must learn incrementally on the input stream of sensor readings and user supervision. In this work, we: (i) introduce interactive classification in the wild and present knowledge drift (KD), a special form of concept drift, occurring due to world and user changes; (ii) develop simple and robust ML methods to tackle these scenarios; (iii) showcase the advantages of each of these methods in empirical evaluations on controlled synthetic and real-world data sets; (iv) design a flexible and modular architecture that combines the methods above to support context recognition in the wild; (v) present an evaluation with real users in a concrete social science use case."]},{"key":"dc:title","label":"Title","values":["Human-Machine Alignment for Context Recognition in the Wild"]}]}],"canonical_facts":{"dc:contributor":["Bontempelli, Andrea","Giunchiglia, Fausto","Passerini, Andrea"],"dc:creator":["Bontempelli, Andrea"],"dc:date":["2024-04-30"],"dc:description":["The premise for AI systems like personal assistants to provide guidance and suggestions to an end-user is to understand, at any moment in time, the personal context that the user is in. The context – where the user is, what she is doing and with whom – allows the machine to represent the world in user’s terms. The context must be inferred from a stream of sensor readings generated by smart wearables such as smartphones and smartwatches, and the labels are acquired from the user directly. To perform robust context prediction in this real-world scenario, the machine must handle the egocentric nature of the context, adapt to the changing world and user, and maintain a bidirectional interaction with the user to ensure the user-machine alignment of world representations. To this end, the machine must learn incrementally on the input stream of sensor readings and user supervision. In this work, we: (i) introduce interactive classification in the wild and present knowledge drift (KD), a special form of concept drift, occurring due to world and user changes; (ii) develop simple and robust ML methods to tackle these scenarios; (iii) showcase the advantages of each of these methods in empirical evaluations on controlled synthetic and real-world data sets; (iv) design a flexible and modular architecture that combines the methods above to support context recognition in the wild; (v) present an evaluation with real users in a concrete social science use case."],"dc:identifier":["https://hdl.handle.net/11572/407669","http://dx.doi.org/10.15168/11572_407669","10.15168/11572_407669"],"dc:language":["eng"],"dc:publisher":["Università degli studi di Trento","place:TRENTO"],"dc:relation":["firstpage:1","lastpage:139","numberofpages:139"],"dc:rights":["info:eu-repo/semantics/openAccess","license:Creative commons","license uri:http://creativecommons.org/licenses/by-sa/4.0/"],"dc:subject":["interactive machine learning, incremental learning, data streams human-in-the-loop, concept drift, context recognition"],"dc:title":["Human-Machine Alignment for Context Recognition in the Wild"],"dc:type":["info:eu-repo/semantics/doctoralThesis"]},"updated_at":"2026-07-24T05:04:24Z"}