Title: ScenarioSA: A Large Scale Conversational Database for Interactive Sentiment Analysis

Abstract: Interactive sentiment analysis is an emerging, yet challenging, subtask of the sentiment analysis problem. It aims to discover the affective state and sentimental change of each person in a conversation. Existing sentiment analysis approaches are insufficient in modelling the interactions among people. However, the development of new approaches are critically limited by the lack of labelled interactive sentiment datasets. In this paper, we present a new conversational emotion database that we have created and made publically available, namely ScenarioSA. We manually label 2,214 multi-turn English conversations collected from natural contexts. In comparison with existing sentiment datasets, ScenarioSA (1) covers a wide range of scenarios; (2) describes the interactions between two speakers; and (3) reflects the sentimental evolution of each speaker over the course of a conversation. Finally, we evaluate various state-of-the-art algorithms on ScenarioSA, demonstrating the need of novel interactive sentiment analysis models and the potential of ScenarioSA to facilitate the development of such models.