| Excerpt |
|---|
Webist 2014 Conference
We attended to the Webist 2014 conference with two papers: "A Survey on Challenges and Methods in News Recommendation" - Özlem Özgöbek, Jon Atle Gulla and R. Cenk Erdur "Implicit User Profiling in News Recommender Systems" - Jon Atle Gulla, Arne Dag Fidjestøl, Xiaomeng Su and Humberto Castejon Data CollectionReal systems work on data and to be able to build effective systems we need realistic data sets. Without the proper data set it is not possible to develop, test and improve the systems. Currently, there is no proper open data set for news domain and Norwegian news. So we are going to build a data set specific to news domain. We are calling for participation of data collection. Call for data collection. (PDF) NRA 2014International Workshop on News Recommendation and Analytics (NRA 2014) will be held in conjunction with 22nd Conference on User Modelling, Adaptation and Personalization (UMAP 2014) on 7-11 July 2014 // Aalborg, Denmark. Topics of interests for this workshop include but are not limited to:
|
Telenor-IDI Seminar
The seminar is held on 12 December 2013 in collaboration of Telenor and IDI. The main focus of the seminar was about the current progress of the SmartMedia project and recommender systems. You can find the program and presentations given during the seminar in here.
NxtMedia Conference
In 8-9 October 2013 we participated the seminar at NxtMedia Conference about the potential of recommendation technology.
NRS 2013
We are delighted to organise the first International Workshop and Challenge on News Recommender Systems at the 7th ACM Recommender System Conference (RecSys 2013) in Hong Kong. This workshop intends to bring together researchers and practitioners around the topics of designing and evaluating novel news recommender systems in order to: (1) share research on news recommendation techniques and evaluation methodologies (2) explore key challenges in the area, and (3) identify emerging topics. Additionally, a challenge will allow participants to evaluate their method by directly interacting with a real-world news recommender systems. The challenge will feature a data set designed to bootstrap a news recommender system. During the weeks before the conference, we will record how well each participant’s system performs with respect to the ratio of clicks per recommendation request. The observed performances will be outlined. We will award the best performing systems. This workshop aims at creating an interdisciplinary community with a focus on the design issues for news recommender systems and promoting the collaboration opportunities between researchers and practitioners.
The workshop will be held a sequence of talks structured as sessions. Such sessions will allow participants to present their work. We reserve one session to present the challenge’s results. Participants who took part in the challenge will have the chance to present their method during this session. Besides challenge contributions, we welcome submissions covering topics related to news recommender systems. Such topics include but are not limited to:
Topics of interest
- Recommendation techniques
- Tracking of news evolution
- Evaluation approaches
- The interplay of news and social-media data
- User interface issues
- News mining and analytics
- News reader behavioural models
- Semantic and news context analysis
- User profiling and preference elicitation
- News recommendations on mobile platforms
- Information retrieval in news collections
Demonstration at WWW 2013
Our demo paper is accepted at WWW 2013.
Widget Connector url http://youtu.be/3HgvnlqZ67A