Search method and system based on intelligent agent knowledge base

Abstract

The invention discloses a search engine based on an intelligent agent knowledge base, which comprises the following steps: inputting search content, and carrying out word segmentation on the search content; performing filtering and error correction on the segmented words and then performing semantic analysis to obtain semantic information of the segmented words; performing matching query on the semantic information and multiple groups of questions in a knowledge base, and if the semantic information is matched with the multiple groups of questions, outputting answers corresponding to one groupof questions to be matched; and if not, performing matching query on the semantic information in the knowledge base through full-text retrieval, and outputting answers corresponding to a group of questions to be matched. The search efficiency and accuracy can be improved in the mode that natural semantic retrieval is combined with full-text retrieval.

Description

technical field [0001] The invention relates to the technical field of search engines, in particular to a search engine and system based on an intelligent agent knowledge base. Background technique [0002] With the rapid development of the Internet in recent years, the number of people surfing the Internet has shown a trend of rapid growth. As the largest and most convenient information carrier, the Internet has become an important way for many people to acquire knowledge. [0003] However, the amount of information on the Internet is growing explosively. A few years ago, the amount of web pages included in global search engines was only tens of millions of pages, but now it has reached billions of pages. The increase in the number has brought about the popularity of search services, and at the same time, the quality of traditional comprehensive searches, such as Baidu and Google, has declined. Search engines are far behind the growth rate of network information in collect...

AI Technical Summary

Problems solved by technology

The increase in the number has brought about the popularity of search services, and at the same time, the quality of traditional comprehensive searches, such as Baidu and Google, has declined. Search engines are far behind the growth rate of network information in collecting network information. Yes, there are often hundreds of thousands of data, which look good, but it is difficult to find suitable results. There are a lot of repetitive information and spam in the search results, and it is difficult for users to accurately filter out the desired results in a short time. Content

[0004] At present, in the operation process of most companies’ call centers, the knowledge base used by agents has the following problems: the knowledge structure is single, mainly unstructured knowledge, the knowledge search is not accurate, the response is slow, and the user needs to have some knowledge about the search. You must understand that the search lacks knowledge correlation, and the knowledge display form is single. Customer service cannot quickly find knowledge in the knowledge base, and quickly learn and master knowledge through the knowledge base.

Therefore, the existing search technology cannot accurately locate knowledge, and requires users to have a certain understanding of the searched knowledge, which has disadvantages such as low efficiency of knowledge search, no system of knowledge accumulation, and little knowledge connection.

Images