Current Projects

AHTOM: Asynchronous and Heterogenous Topic Mining

The project will develop a set of models and respective software tools for topic mining of asynchronous and heterogeneous streams that will enable: (a) the discovery of a set of topics expressed by heterogeneous documents in multiple streams; (b) the quantification of the similarity between documents in terms of content re-use; (c) the tracking of the flow and changes of content over time; and (d) the discovery of significant temporal dependencies among the streams, providing quantitative measures of the importance, authoritativeness and timeliness of each stream in relation to a topic or set of topics.

Funding Agency: Secretariat d'Etat a l'Education et a la Researche (SER)
Starting date: Jan 2016 
Duration: 24 months

RelMobIR (RELevance criteria combination for MOBile Information Retrieval)

Context-aware mobile IR systems are able to estimate the relevance of a given page in relation to several different aspects. The first aspect is obviously topical relevance, but other types of relevance play an important role, like for example geographic, temporal or opinion relevance.  This project addresses the problem of combining relevance criteria for mobile IR, in order to determine the effective overall relevance of a document to a user information need. The work starts by addressing how to best estimate these different relevance criteria for a document, given what is know of the user’s opinions, preferences and his spatiotemporal context. It then addresses the problem of the best combination of these relevance estimates, under the assumption of their independence from each other.

Funding Agency: Swiss National Science Foundation (SNSF)
Starting date: April 2015 
Duration: 36 months

OpiTrack: Tracking Opinion Changes over Time

Recent years have seen the rapid growth of social media platforms that enable people to express their thoughts and opinions on the web and share them with other users. Many people write their opinion about products, movies, people or events on blogs, forums or review sites. In OpiTrack we will extend the latest models of opinion retrieval and sentiment analysis by incorporating time in the opinion analysis. We will develop models that track sentiment over time also covering different aspects of a topic. We will also develop models that will attempt to predict the sentiment change toward a topic at a certain time in future. The models developed will be evaluated with real data.

Funding Agency: Swiss National Science Foundation (SNSF)
Starting date: June 2014 
Duration: 36 months


The RECALL project aims to re-think and re-define the notion of memory augmentation. Recent developments in capture technology and information retrieval allow for continuous and automated recordings of many aspects of our everyday lives. RECALL will harness these trends and develop a new paradigm for memory augmentation technologies that are technically feasible, desired by users, and beneficial to society.

Funding Agency: EU FET
Starting date
: Nov 2013 
Duration: 36 months

KEYSTONE: Semantic Keyword-based Search on Structured Data Sources

Keyword searching is becoming a valuable alternative to traditional structured queries, mainly due to its simplicity and the basic knowledge required from users. Nevertheless, existing approaches suffer from a number of limitations when applied to multi-source scenarios requiring some form of query planning, and with frequent updates precluding any effective implementation of data indexes. KEYSTONE proposes to join synergies from several disciplines, such as semantic data management, the semantic web, information retrieval, artificial intelligence, machine learning, user interaction, service science, service design, and natural language processing, to build an effective solution to this problem.

Funding Agency: Information and Communication Technologies (ICT) Action IC1302 Starting date: May 2013
Duration: 48 months