企业微信_客服违禁语识别能力

对企业微信会话文本/句子进行违禁语识别，通过大数据平台收集企业微信会话数据，利用业务规则与深度学习方法训练违禁语识别模型，将其封装成为API。实现对客服会话内容涉及到的违禁语、敏感话题、脏话等进行识别，提升一线服务人员话术的合规性。This system targets prohibited language detection for WeChat Work chat texts and sentences. Specifically, WeChat Work chat data is collected via big data platforms first. Then, a prohibited language detection model is trained using both business rules and deep learning methods, and the trained model is encapsulated into an API. The system is capable of identifying prohibited language, sensitive topics, vulgar language and other non-compliant content within customer service chat sessions, thereby improving the compliance of frontline service personnel's communication wording.

搜集汇总

背景与挑战

背景概述

该数据集基于企业微信会话数据，采用规则与深度学习结合的方式训练违禁语识别模型，用于检测客服对话中的敏感内容以提升服务合规性。