Multi-Turn Dialogue Generation in E-Commerce Platform with the Context of Historical Dialogue

Findings of the Association for Computational Linguistics: EMNLP 2020, pages 1981–1990 November 16 - 20, 2020. c⃝2020 Association for Computational Linguistics 1981 Multi-Turn Dialogue Generation in E-Commerce Platform with the Context of Historical Dialogue Weisheng Zhang∗1, Kaisong Song∗2, Yangyang Kang2, Zhongqing Wang†1, Changlong Sun2, Xiaozhong Liu2, Shoushan Li1, Min Zhang1, Luo Si2 1School of Computer Science and Technology, Soochow University, China 2Alibaba Group, Hangzhou, Zhejiang, China wszhang0@stu.suda.edu.cn, {kaisong.sks, yangyang.kangyy, luo.si}@alibaba-inc.com, {wangzq, lishoushan, minzhang}@ suda.edu.cn, changlong.scl@taobao.com, liu237@indiana.edu Abstract As an important research topic, customer ser- vice dialogue generation tends to generate generic seller responses by leveraging current dialogue information. In this study, we pro- pose a novel and extensible dialogue gener- ation method by leveraging sellers’ historical dialogue information, which can be both ac- cessible and informative. By utilizing inno- vative historical dialogue representation learn- ing and historical dialogue selection mecha- nism, the proposed model is capable of detect- ing most related responses from sellers’ his- torical dialogues, which can further enhance the current dialogue generation quality. Un- like prior dialogue generation efforts, we treat each seller’s historical dialogues as a list of Customer-Seller utterance pairs and allow the model to measure their different importance, and copy words directly from most relevant pairs. Extensive experimental results show that the proposed approach can generate high- quality responses that cater to specific sellers’ characteristics and exhibit consistent superior- ity over baselines on a real-world multi-turn customer service dialogue dataset. 1 Introduction Over the past years, online shopping has expe- rienced incredible growth. In e-commerce plat- forms, e.g., Amazon and Taobao, brilliant cus- tomer service is becoming increasingly important because of significantly reducing the workload of shop sellers. Ideally, sellers should provide high- quality responses to address the personal needs of the customers. However, such cost can be pro- hibitive for most small businesses, which inspires us to be concerned with the multi-turn dialogue ∗Both authors contributed equally to this research. †Corresponding Author: Zhongqing Wang. Current Dialogue Customer Hello. Seller I’m grad to service you, dear. Seller What can I do for you? Customer 1.65 meters tall and weigh 48 kg, which size should I buy? Seller In my experience, you may fit the M size. Customer Aright, when could you send it off? Seller As soon as we can. Customer Will you give me some discount? Seller’s Historical Dialogues C1 Hello. C2 I’m looking for some help. S1 Welcome to our store. S2 What can I do for you? C3 I see and is there any coupons? S3 You can find it on our main page. S4 Click the link to get it. C4 OK, I find it, thank you. S5 I’m grad that I can help you. Seller’s Response HRED I’m sorry not. Ground Truth You can open the main page of our store and draw a coupon. Table 1: The example of customer server dialogue be- tween the Seller (S) and the customer (C) plus the gen- erative results. The above block is the current dialogue context, the middle one is …

Content truncated to 500 words