Extractive Summarization Method for Contact Center Dialogues based on Call Logs

This paper proposes a novel extractive summarization method for speech dialogues between agents and customers in contact centers. The proposed method does not require any extra cost for applying the method such as preparing rules or creating training data. Conventional methods such as the tf*idf method, which gives importance to characteristic words in an input text, can miss the essential points for contact center work. Our proposed method evaluates the importance of each utterance from the standpoint of call agents who report calls for managing or analyzing calls. Specifically, the proposed method includes information frequently reported by call agents in summaries using past call logs commonly recorded in the contact center. Evaluation using real data (call dialogues and call logs) shows that the proposed method can extract essential points in terms of contact center work and outperforms the conventional method.