Preset parsing method for tourism vertical search engine

Abstract

The invention discloses a preset parsing method for a tourism vertical search engine. The method can analyze search words of a user, conduct semantic analysis, further combine rules after matching with a plurality of rule engines, and eventually convert the search words to query conditions that meet user's requirements. The method of the invention can conduct semantic parsing based on search word text, understand the intent of user's search words, map an original request to query conditions that are more suitable for user's requirements, and make query result more accurate. The flexible multi-rule engine application method of the invention facilitates developers self-defining more rule engine and optimizing search results. After a rule database is modified, the preset parsing automatically identifies modification content through timestamp, and the preset parsing service is dynamically updated, and the rule engine is updated through a newly opened memory space, so that the preset parsing service is not interrupted, which facilitates operations personnel rapidly adjusting search results based on key words without performing development and code issuing, thereby substantially increasing operation efficiency.

Description

technical field [0001] The invention belongs to the technical field of search engines, and in particular relates to a pre-analysis method for vertical search engines of tourism. Background technique [0002] With the development of Internet tourism platforms, the tourism products that can be sold online have been greatly enriched. How to let users quickly find the products they want has become a key problem for major tourism websites to solve. The travel vertical search engine can facilitate users to quickly find products, and is an indispensable quick search tool. [0003] How to return the product that the user wants is a difficult problem in the vertical search engine. The general vertical search engine returns the product by matching the text of the search term, for example, considering the word frequency of the search keyword in the product. However, when the word frequency cannot represent the matching correlation, products that the user does not want may be returned....

AI Technical Summary

Problems solved by technology

However, when the word frequency cannot represent the matching correlation, it may return products that the user does not want. For example, the user searches for "Maldives" and may return the Sanya tourism product of "Little Maldives".

At the same time, when the search results of certain search terms are not ideal, it is difficult for general vertical search engines to quickly adjust the results, which leads to a large deviation between the retrieval results of existing vertical search engines and the needs of users

Images