According to a recent claim by IBM, 90% of the data available today have been created in the last two years. This exponential growth of online information gave new life to research in the area of user modelling and personalization, since information about usersí preferences, sentiment and opinions as well as signals describing their physical and psychological state can now be obtained by mining data gathered from many heterogeneous sources.
Such sources can be roughly classified into two categories: on one side, the recent trend of Quantified Self (QS) and Personal Informatics emphasized the use of technology to collect personal data on different aspects of peopleís daily lives. These data can be internal states (such as mood or glucose level) or indicators of performance (such as the kilometres run). The purpose of collecting these data is self-monitoring, performed to gain self-knowledge or to obtain some change or improvement (behavioural, psychological, therapeutic, etc.). Often these data are also exploited for behaviour change purposes, for example to increase the userís physical activity.
At the same time, an enormous amount of textual content is continuously spread on social networks, and this has driven a strong research effort to investigate to what extent such data can be exploited to infer user interests, personality traits, emotions, and knowledge. Moreover, the recent phenomenon of (Linked) Open Data fuelled this research line by making available a huge amount of machine-readable textual data that can be used to connect all the data points spread in different data silos under a uniform representation formalism.
The main research questions, which arise from both trends, are quite fundamental: how can we separate signal from noise and extract some real value from this plethora of data produced through devices and social networks? How can we effectively merge such data to obtain a holistic (and semantic) representation of all the facets describing people? How can we use such data to trigger personalization and adaptation mechanisms? Are there new opportunities for providing extremely tailored services for supporting users in their everyday life? What about the privacy issues, the need for transparency and the ethical implications in the management of such data?
The workshop aims to provide a forum for discussing open problems, challenges and innovative research approaches in the area, in order to investigate whether the adoption of techniques for semantic representation of textual and physiological data points can be effective to build a new generation of personalized and intelligent systems based on the analysis of Personal, Big and Linked Open Data.
Topics of interests include but are not limited to:
Personal, Linked and Social Data Mining
– Techniques for collection, aggregation and analysis of Personal, Linked and Social Data
– Social Sensing (aggregating user-based data to obtain people-based findings)
– Scalability issues and technologies for massive social data extraction
– Ethical issues, need for transparency, privacy management of Personal and Social data
– Semantic Holistic User Profiling and EU General Data Protection Regulation (GDPR)
Semantic Analysis of Personal, Linked and Social data
– Semantics Representation based on Open Knowledge Sources (Wikipedia, DBpedia, Freebase, etc.)
– Semantics Representation based on Entity Linking algorithms (TagMe, Babelfy, DBpedia Spotlight)
– Semantics Representation based on Geometrical Models (e.g. Distributional Models, Deep Learning approaches)
– User Modeling based on Semantic Content Analysis of Social and Linked Open Data
– User Modeling based on data coming from self-tracking devices
– User Modeling based on Emotions and Personality Traits
– Tracking implicit feedbacks (e.g. social activities) to infer user interests;
– Lifelogging User Models
– Recommender Systems exploiting Personal and Linked Data
– Recommender Systems based on Emotions and Personality
– Recommender systems based on physiological data
– Recommender systems for behavior change
– Adaptation and Personalization in e-Government domain
– Online Monitoring based on Social Data (Social CRM, Brand Analysis, etc.)
– Intelligent and Personalized Smart Cities-related Applications (e.g. Event Detection, Incident Detection, Personalized Planners, etc.)
We encourage the submission of original contributions, investigating the impact of content analysis techniques on adaptive and personalized services:
(A) Full research papers (max 5 pages – ACM format);
(B) Short Research papers and Demos (max 2 pages – ACM format);
Submission site: https://easychair.org/conferences/?conf=hum2017
All submitted papers will be evaluated by at least two members of the program committee, based on originality, significance, relevance and technical quality. Papers should be formatted according to the ACM SIG proceedings template: http://www.acm.org/publications/proceedings-template
Note that the references do not count towards page limits. Submissions should be single blinded, i.e. authors names should be included in the submissions.
Submissions must be made through the EasyChair conference system prior the specified deadline (all deadlines refer to GMT). At least one of the authors should register and take part at the conference to make the presentation.
All accepted papers will be published by ACM as a joint volume of Extended UMAP 2017 Proceedings and will be available via the ACM Digital Library. At least one author of each accepted paper must register for the particular workshop and present the paper there.
* Full paper submission: April 27, 2017 (EXTENDED!)
* Paper notification: May 20, 2017
* Camera-ready paper: May 28, 2017
Cataldo Musto – University of Bari, Italy
Amon Rapp – University of Torino, Italy
Federica Cena – University of Torino, Italy
Frank Hopfgartner – University of Glasgow
Judy Kay – University of Sydney, Australia.
Giovanni Semeraro – University of Bari, Italy