Semantic sharing platform system

Abstract

The invention relates to a semantic sharing platform system which comprises a word bank and database construction system, a self-learning engine system and a customer service assistant application system. The self-learning engine system comprises an intelligent recommendation module and an intelligent maintenance and detection module; the intelligent recommendation module is used for realizing the functions of automatic clustering, automatic pushing and automatic data statistics on text data meeting hot event conditions, automatically judging whether a customer is interested in similar knowledge points or not according to historical consultation, and pushing information contents of similar problems to a service system; the intelligent maintenance and detection module is used for automatically aggregating similar questions according to a sample set of customer questions to improve the semantic maintenance efficiency; when the training personnel maintain the semantic knowledge, a conflict reminding function is provided; when the word model is maintained, conflicting semantics of the semantic platform are automatically judged, and manual intervention is carried out to improve the precision of semantic recognition.

Description

technical field [0001] The present invention relates to a semantic sharing platform system, more specifically, relates to a semantic sharing platform system of a multimedia customer service platform. Background technique [0002] During the online-to-offline service, when the user wants to search for a place of interest, the user can input location-related query words into the query box of the online and offline service application implemented on the user terminal device, The online-to-offline system can recommend a location to the user based on the semantics of the input query words. At least one recommended location can help the user to quickly find the item that the user wants. In the existing methods for semantic search, the online-to-offline system can parse the query words and go through at least two recall strategies to retrieve at least one recommended location for the user, however, the problem with the existing methods is that the search strategy Complicated, resu...

AI Technical Summary

Problems solved by technology

In the existing methods for semantic search, the online-to-offline system can parse the query words and go through at least two recall strategies to retrieve at least one recommended location for the user, however, the problem with the existing methods is that the search strategy Complicated, resulting in low search efficiency and high search costs