A kind of information processing method and computer equipment

Abstract

The present application relates to an information processing method and computer equipment. After determining each candidate instruction information corresponding to multiple candidate intents of a question sentence, the method does not directly return and display each candidate instruction information associated with the corresponding answer content, but It is further based on the association relationship between each candidate indication information to generate the final target indication information for display (of course, the corresponding answer content will be associated with each target indication information), which can undoubtedly make the final target indication information displayable. Provide users with more information in a format that can reflect the correlation between different candidate indication information, thereby providing help for users to understand the indication information, and can effectively reduce the user's understanding of the answer indication information and answer indication. The deviation between the business knowledge and semantic meaning actually represented by the information improves the human-computer interaction efficiency of the intelligent customer service system and the user's trust / satisfaction with the intelligent customer service system.

Description

technical field [0001] The present application belongs to the technical field of artificial intelligence, and in particular relates to an information processing method and computer equipment. Background technique [0002] In the intelligent question and answer scenario based on intelligent customer service, for user questions, the intelligent customer service system will generally first identify the user's intent embodied in the user's question, and then generate an answer corresponding to the user's question based on the identified intent information. Usually, each user question corresponds to multiple candidate intents. When generating an answer based on an intent, for each candidate intent, corresponding indication information (such as an answer title) and answer content are generally generated, where the indication information is used for conciseness Briefly reflect the central idea and main content of the answer. [0003] For a friendly interface, when a user asks a qu...

AI Technical Summary

Problems solved by technology

In some cases, for example, the user's question or scene is relatively complicated, or the user's understanding of business knowledge is insufficient, etc., the user may not fully understand the actual meaning of the instruction information returned by the system, that is, the instruction understood from the user's perspective The business knowledge and semantic content actually represented by the information and the instruction information may not be completely consistent, which may easily lead to the fact that even if the system understands the user's intention correctly and returns the correct answer, the user's understanding of the instruction information may not be implemented due to deviations in the user's understanding of the instruction information. The correct answer leads to the fact that the user's problem is not really solved, which reduces the human-computer interaction efficiency of the intelligent customer service system, and also reduces the user's trust / satisfaction with the intelligent customer service system

Images