Abstract de la publi numéro 17781

In this paper, we specifically consider the challenging task of solving a question posted on Twitter. The latter gener- ally remains unanswered and most of the replies, if any, are only from members of the questioner's neighborhood. As outlined in previous work related to community Q&A, we believe that question-answering is a collaborative process and that the relevant answer to a question post is an aggre- gation of answer nuggets posted by a group of relevant users. Thus, the problem of identifying the relevant answer turns into the problem of identifying the right group of users who would provide useful answers and would possibly be will- ing to collaborate together in the long-term. Accordingly, we present a novel method, called CRAQ, that is built on the collaboration paradigm and formulated as a group en- tropy optimization problem. To optimize the quality of the group, an information gain measure is used to select the most likely “informative” users according to topical and col- laboration likelihood predictive features. Crowd-based ex- periments performed on two crisis-related Twitter datasets demonstrate the effectiveness of our collaborative-based an- swering approach.