Overview of NTCIR-13 ECA Task

This paper presents the overview of NTCIR-13 emotion cause analysis (ECA) task. This task is designed to evaluate the emotion cause analysis techniques which automatically determine whether a clause contain the cause of emotion expression. Unlike the regular evaluation tasks on sentiment analysis or emotion classification, this evaluation task focus on exploring the cause stimulate the emotion. To support this evaluation, the text are manually labelled emotion category and emotion cause to construct an emotion cause corpus which follows the scheme of W3C Emotion Markup Language. We designed two subtasks: The emotion cause detection subtask for identifying the clause containing the emotion cause corresponding to an emotion expression. The emotion cause extraction subtask is designed to evaluate the techniques for detecting the exact boundary of the emotion cause. Eleven teams participated this task and eventually, three valid results for emotion cause detection subtask on Chinese side are submitted. Evaluations shows that the top system achieved 66.5% F-score.