INEX Tweet Contextualization Task: Evaluation, Results and Lesson Learned

Patrice Bellot (1) , Véronique Moriceau (2) , Josiane Mothe (3, 4) , Eric San Juan (5) , Xavier Tannier (2, 6, 7)

1 LSIS - Laboratoire des Sciences de l'Information et des Systèmes
2 LIMSI - Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur
3 IRIT-SIG - Systèmes d’Informations Généralisées
4 UT2J - Université Toulouse - Jean Jaurès
5 AU - Avignon Université
6 Université Paris-Saclay
7 UP11 - Université Paris-Sud - Paris 11

Microblogging platforms such as Twitter are increasingly used for on-line client and market analysis. This motivated the proposal of a new track at CLEF INEX lab of Tweet Contextualization. The objective of this task was to help a user to understand a tweet by providing him with a short explanatory summary (500 words). This summary should be built automatically using resources like Wikipedia and generated by extracting relevant passages and aggregating them into a coherent summary. Running for four years, results show that the best systems combine NLP techniques with more traditional methods. More precisely the best performing systems combine passage retrieval, sentence segmentation and scoring, named entity recognition, text part-of-speech (POS) analysis, anaphora detection, diversity content measure as well as sentence reordering. This paper provides a full summary report on the four-year long task. While yearly overviews focused on system results, in this paper we provide a detailed report on the approaches proposed by the participants and which can be considered as the state of the art for this task. As an important result from the 4 years competition, we also describe the open access resources that have been built and collected. The evaluation measures for automatic summarization designed in DUC or MUC were not appropriate to evaluate tweet contextualization, we explain why and depict in detailed the LogSim measure used to evaluate informativeness of produced contexts or summaries. Finally, we also mention the lessons we learned and that it is worth considering when designing a task.

Mots clés: Short text contextualization; tweet contextualization; tweet understanding; automatic summarization; contextual information retrieval; question answering; focus information retrieval; natural language processing; Wikipedia; text readability; text informativeness; textual references; Kullback-Leibler divergence

Domaines: Informatique [cs]

Patrice Bellot, Véronique Moriceau, Josiane Mothe, Eric San Juan, Xavier Tannier. INEX Tweet Contextualization Task: Evaluation, Results and Lesson Learned. Information Processing and Management, 2016, 52 (5), pp.801-819. ⟨10.1016/j.ipm.2016.03.002⟩. ⟨hal-01479297⟩