=Paper=
{{Paper
|id=Vol-1618/IFUP_preface
|storemode=property
|title=None
|pdfUrl=https://ceur-ws.org/Vol-1618/IFUP_preface.pdf
|volume=Vol-1618
}}
==None==
https://ceur-ws.org/Vol-1618/IFUP_preface.pdf
Preface: Papers and Research from the 2016 International
Workshop IFUP
Guibing Guo Robin Burke
Software College, DePaul University, US
Northeastern University, China rburke@cs.depaul.edu
guogb@swc.neu.edu.cn
Feida Zhu Neil Yorke-Smith
Singapore Management American University of Beirut,
University, Singapore Lebanon
fdzhu@smu.edu.sg nysmith@aub.edu.lb
We are pleased to introduce a set of papers from the 2016 citing ideas and research work can be inspired from these
international workshop IFUP held in in Halifax, Canada on papers, which is the main objective of our workshop. For
16 July 2016. These papers focus on the two applications the research lines of future series of our workshop IFUP, we
of multi-dimensional information fusion, i.e., user modelling would like to encourage more research efforts and attention
and personalization. In total, six quality papers are accept- to the following topics.
ed this year, each of which received three reviews from the
IFUP 2016 program committees. Twelve PC members and • Heterogenous feedback based recommendation. Con-
workshop chairs, spreading over 9 countries and institutes, ventionally, researchers tend to split the user feedback
contribute a lot to the success of our workshop. into explicit and implicit feedback, and only focus on
Multi-dimensional information fusion is a challenge topic either type of feedback in their work, ignoring the
in both user modelling and personalization, especially giv- possible existance of the other kind of feedback. It
en the ever-growing amount of information as well as the is necessary to have more thoughts on how different
number of information types. Cena et al. take an empirical types of interactions may interplay with each other.
study in user models based on cross-representation media- In this regard, temporal information or feedback se-
tion. Other researchers are more in favor of the application quences could be a very useful and indicative informa-
in recommender systems, i.e., personalization. Auxiliary in- tion source to help resolve such an issue.
formation (e.g., social connections, item category and de-
• Online and offline information differentiation and inte-
scription) has been incorporated in many recommendation
gration. Online user behaviors can have distinct char-
models to enhance the performance of two general recom-
acteristics from offline user behaviors. A user may have
mendation tasks, namely rating prediction and item rank-
many online friends but only few offline contacts. On-
ing. Specifically, Alotaibi and Vassileva attempt to combine
line behaviors may have less constraints than offline
both explicit and implicit social networks for personalized
ones, and the rules and regulation of online behaviors
recommendation. Kamehkhosh et al. construct a track mu-
can also have different content from those of offline be-
sic recommender system with the consideration of multi-
haviors. For the applications (e.g., mobile Apps) ori-
dimensional long-term preference. Peng et al. propose a
enting to different types of users (i.e., online, offline, w-
RBPR method to better rank items in the context of het-
hole), it is necessary to distinguish one from the other,
erogeneous implicit feedback. Some others are concerned
or integrate both together for the whole-view picture.
with the use of additional information. For example, Liang
et al. take into consideration social trust to resolve the cold • More understanding of items. So far, we have noted
start problem and thus improve the performance of rating many research papers surfaced in the field of social
prediction. Chen et al. contend that time is an important recommender systems in the light of user-user associ-
factor for collaborative filtering. ations. However, it seems a lack of research to better
Together, the body of research work of IFUP 2016 have understand the associations among items. Along with
taken an initial step to boost the research and application the development of new applications and technologies
of multi-dimensional information fusion. We hope more ex- (e.g., IoT), it may be easier to extract and infer many
kinds of item-item relationships, such as complementa-
tion, co-existance, mutual exclusion, etc. Leveraging
these relationships can be helpful in improving both
the accuracy and explanability of recommendations.
Our workshop IFUP will also look into other research di-
rections regarding multi-dimensional information fusion.