arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.04668cs.CLcs.AIcs.IR
收藏

从浏览历史预测在线客服对话意图

Forecasting Live Chat Intent from Browsing History

University of California, San Diego(加利福尼亚大学圣迭戈分校);Lowe’s Companies, Inc.(劳氏公司)

Se-eun Yoon, Ahmad Bin Rabiah, Zaid Alibadi, Surya Kallumadi, Julian McAuley

AI总结:

本文提出一种两阶段方法,先利用微调Transformer对浏览历史进行高层意图分类,再结合LLM生成细粒度用户意图,并通过LLM自动评估验证了该方法的显著性能优势。

摘要:

客户带着各种意图联系在线客服人员,例如询问产品详情或请求退货。在本文中,我们提出了从浏览历史预测用户意图的问题,并通过两阶段方法加以解决。第一阶段将用户的浏览历史分类为高层意图类别。在此阶段,我们将每条浏览历史表示为页面属性的文本序列,并使用真实类别标签对预训练的Transformer模型进行微调。第二阶段向大语言模型(LLM)提供浏览历史和预测的意图类别,以生成细粒度意图。在自动评估方面,我们使用一个独立的LLM来判断生成意图与真实意图之间的相似度,该评估结果与人类判断高度一致。与不经过分类阶段直接生成意图相比,我们的两阶段方法取得了显著的性能提升。

英文摘要:

Customers reach out to online live chat agents with various intents, such as asking about product details or requesting a return. In this paper, we propose the problem of predicting user intent from browsing history and address it through a two-stage approach. The first stage classifies a user's browsing history into high-level intent categories. Here, we represent each browsing history as a text sequence of page attributes and use the ground-truth class labels to fine-tune pretrained Transformers. The second stage provides a large language model (LLM) with the browsing history and predicted intent class to generate fine-grained intents. For automatic evaluation, we use a separate LLM to judge the similarity between generated and ground-truth intents, which closely aligns with human judgments. Our two-stage approach yields significant performance gains compared to generating intents without the classification stage.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑