Method and device for session-based recommendation

Abstract

The invention aims to provide a method and device for session-based real-time recommendation. Based on a current session between a current user and his current customer, according to the current customer's attribute information, a recommendation answer corresponding to a current question in the current session is obtained by matching from a knowledge base; and the recommendation answer is providedto the current user; wherein the structure of knowledge in the knowledge base includes customer attributes, questions and answers. Compared with the prior art, the method and device for session-basedreal-time recommendation cause existing knowledge to be effectively utilized, the current user answer the current customer's query conveniently, and the user's use experience is improved.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a technique for recommending based on sessions. Background technique [0002] For departments or companies that require a large number of voice calls, they have accumulated a large amount of voice materials, which include a lot of experience and knowledge. However, due to the lack of effective management, it is difficult for new employees to benefit from it. A major problem is that manual speech translation requires a lot of manpower and material resources, and it is not easy for new employees to master the extracted knowledge proficiently. [0003] Existing solutions mainly rely on manual knowledge extraction, and rely on users to memorize and apply them. However, the existing methods have the following disadvantages: 1) Manual extraction is difficult to ensure the comprehensiveness of knowledge, and it depends entirely on personal experience. Knowledge is ignored, o...

AI Technical Summary

Problems solved by technology

But due to the lack of effective management, it is difficult for new employees to benefit from it

A major problem is that manual speech translation requires a lot of manpower and material resources, and it is not easy for new employees to master the extracted knowledge proficiently

[0003] Existing solutions mainly rely on manual knowledge extraction, and rely on users to memorize and apply them. However, the existing methods have the following disadvantages: 1) Manual extraction is difficult to ensure the comprehensiveness of knowledge, and it depends entirely on personal experience. Knowledge is ignored, or bad knowledge is extracted, so it is difficult to guarantee the optimal answer; 2) The memory of knowledge takes time, which increases the workload of users, and may not be able to remember the correct knowledge at the right time; Knowledge is optimized, good knowledge cannot get more exposure, and bad knowledge will not reduce the number of exposures

Images