Sirris Attended BNAIC 2011

A delegation from the Sirris Software and ICT group attended the 23rd Benelux Conference on Artificial Intelligence (BNAIC 2011), organized by CODeS in Gent on November 2–3, 2011. The conference featured a broad range of both pure and applied research contributions covering topics like machine learning, intelligent agents, operations research, semantic web and data mining.

We presented two contributions:

  • Elena Tsiporkova presented a paper entitled MapReduce and FCA Approach for Clustering of Multiple-Experiment Data Compendium, where she and her colleagues propose a two step method to cluster genes that could possibly share a common biological role. These genes are initially identified through a series of experiments. The proposed method clusters them by first performing a MapReduce operation on each experiment. The resulting clustering solution is further analyzed and refined by performing a formal concept analysis (FCA). You can read the full paper here.
  • Tom Tourwé, Elena Tsiporkova and myself presented a poster paper entitled Ontology-driven Elicitation of Multimodal User Interface Design Recommendations, where we propose an ontology-based modelling framework to capture domain and expert knowledge available within the human-computer interaction (HCI) community. The goal of this framework is to support HCI designers in the selection of the most appropriate modality for interacting with the users of an software-based application, depending on the specific context of these users. Examples of modalities are audio messages and touch input. This research was performed in the context of the ASTUTE and VariBru projects. You can read the full paper here.

Finally, Ruben Van Wanzeele, a student at CODeS, presented the results of his Master's thesis (done under the supervision of Tom Tourwé and Elena Tsiporkova, among others) in a poster paper entitled Extracting emotions out of Twitter's microblogs. In this paper he presents an architecture to automatically extract from the tweets generated during an event the emotions that the authors of these tweets were trying to convey. You can read the full paper here.

Do not hesitate to contact us if you would like to know more about these results!