About LiMoSINe

Posted by Anne Schuth

We increasingly live our life online. Information is accumulated on a wide range of human activities, from science and facts, to personal content, opinions, and trends. Across the globe, people’s knowledge, experiences and interactions effortlessly find their way to online outlets, alongside traditional edited content, ready to be shared with millions. LiMoSINe will integrate the research activities of leading researchers across diverse topics with a view to enabling new kinds of language-based search technology. See about for more information.

Follow us on the following channels:

        

News

Infographic describing RepLab

Posted by admin

We have created an infographic describing RepLab 2013.

CIKM'13 Best Paper Award for "Penguins in Sweaters, or Serendipitous Entity Search on User-generated Content"

Posted by admin

Ilaria Bordino, Yelena Mejova and Mounia Lalmas (all three working on the LiMoSINe Project) won the CIKM'13 Best Paper Award for their paper "Penguins in Sweaters, or Serendipitous Entity Search on User-generated Content". Congratulations!

Mounia Lalmas makes research accessible to all

Posted by admin

Mounia Lalmas from Barcelona Media is writing a blog where she talks about her own research, that of others, in a way that she hopes will make it accessible to all at http://labtomarket.wordpress.com.

Towards the algebraization of Formal Concept Analysis over complete dioids

Valverde-Albacete, F.  J., & Peláez-Moreno C. (2014).  Towards the algebraization of Formal Concept Analysis over complete dioids. Actas del XVII Congreso Español sobre Tecnolog{\'ıas y Lógica Fuzzy, ESTYLF 2014. 93–98.

User Behavior and Bias in Click-Based Recommender Evaluation

Hofmann, K., Schuth A., Bellogin A., & de Rijke M. (2014).  User Behavior and Bias in Click-Based Recommender Evaluation. 36th European Conference on Information Retrieval (ECIR’14).

Optimizing Base Rankers Using Clicks: A Case Study using BM25

Schuth, A., Sietsma F., Whiteson S., & de Rijke M. (2014).  Optimizing Base Rankers Using Clicks: A Case Study using BM25. 36th European Conference on Information Retrieval (ECIR’14).

100 % classification accuracy considered harmful: the normalized information transfer factor explains the accuracy paradox

LiMoSINe Project: Overcoming challenges in Reputation

Posted by admin

In recent years, management of such intangibles as Reputation has been gaining more relevance for companies. Furthermore, the rise of social media brought to companies, institutions, and public figures serious concerns on what is said about them online, as negative mentions can deeply affect businesses or careers. Nowadays, a broad range of specialized consultancies look after their customers, online reputation.

Filtering Relevant Tweets using Probabilistic Signature Models

Anaya-Sanchez, H., Penas A., & Cabaleiro B. (2013).  Filtering Relevant Tweets using Probabilistic Signature Models. CLEF 2013 Evaluation Labs and Workshop. Online Working Notes.