以代理人為基礎之客服電子郵件自動回覆系統

Abstract

企業在面對競爭者的威脅下，如何提升現有客戶的滿意度，並吸引尚未開發的客戶群，一直是企業主管關心的問題。近年來，除了現有的其他溝通方式外，電子郵件已逐漸變成客戶與企業間互動的另一個主要管道。然而，目前大部份的企業仍然採用手動或半自動的方式來回覆客戶的諮詢郵件，客服成本高，因此本研究提出一個以代理人為基礎之客服電子郵件自動回覆系統（CSRAgent），依資訊分類的概念，運用向量空間模型來判別客戶詢問問題的類型，並自動回覆客戶詢問企業之電子郵件。我們亦設計一實驗來驗証本研究所提之架構及方法的正確性，實驗結果顯示，CSRAgent能夠縮短企業回應客戶所需之時間，即時地回覆約80%之諮詢郵件，並確保回函內容之正確性。如此將可為企業降低客服郵件處理之成本，增加生產力，並提升客服滿意度。

As the number of users connected to the Internet continues to skyrocket, e-mail is quickly becoming one of the fastest and most economical forms of communication available. Since customers increasingly communicate with companies using e-mail in place of telephone, the e-mail responding task in most companies remain manual or semi-manual nowadays. Therefore, this thesis develops collaborative e-mail agents, called CSRAgent, capable of automatically replying to mails using a information retrieval model based on the Vector Space Model. Results in this thesis demonstrate that CSRAgent can auto-reply to around 80 percent of mails and decreases mailing response time into a desired accuracy. Consequently, CSRAgent decreases manpower and improves productivity, additionally, and then accelerates the mail replying process, thereby increasing customer satisfaction.