Home // UBICOMM 2017, The Eleventh International Conference on Mobile Ubiquitous Computing, Systems, Services and Technologies // View article
Authors:
Antonios Karatzoglou
Michael Beigl
Keywords: LBS; Location Prediction, Ontologies, Dynamic Semantic Similarity, Personality Traits, Emotional State, Personalization, Cognitive Frames
Abstract:
Location prediction plays a steadily growing role in Location Based Services (LBS), as these try to be more proactive and improve in this way the quality of service provided. Although recent location prediction systems go beyond just location data and build upon a wide range of models that describe semantic locations and personal preferences, none of them considers locations from the view of the user. Moreover, none takes into account the variance in people's way of perceiving and understanding concepts (locations in our case) depending on the situation. Minsky referred to this as (cognitive) emph{frames}. This paper posits that a dynamic semantic-similarity-based clustering of locations can be used for mining such location-specific frames, e.g. the varying meanings that people give to locations over time depending on the situation, their personality and their emotional state. The resulting situation-person-specific frames can in turn be used to enhance the location prediction.
Pages: 95 to 100
Copyright: Copyright (c) IARIA, 2017
Publication date: November 12, 2017
Published in: conference
ISSN: 2308-4278
ISBN: 978-1-61208-598-2
Location: Barcelona, Spain
Dates: from November 12, 2017 to November 16, 2017