Method and device for establishing index database and retrieval method, device and system

Abstract

The invention provides a method and device for establishing an index database and a retrieval method, device and system. The method for establishing the index database comprises the steps that files are preprocessed, and lexical items corresponding to the files are obtained; keywords of the files are calculated, and keywords corresponding to the files are obtained; whether the lexical items are the key words of the files or not is judged, if yes, the reverse index of the lexical items and the corresponding files is established in a first index table, and if not, the reverse index of the lexical items and the corresponding files is established in a second index table. The files highly relative to search words can be rapidly found, the retrieval efficiency is improved while the retrieval accuracy is improved, and a good retrieval experience is provided for a user.

Description

technical field [0001] The present invention relates to the field of natural language processing, in particular to a method, device and retrieval method, device and system for constructing an index library. Background technique [0002] Indexing technology is one of the core technologies of search engines. The search engine organizes, classifies, and indexes the collected information to generate an index library. The method for constructing an index in the prior art includes the following steps: [0003] 1. Preprocess the document, including word segmentation and removing stop words; [0004] 2. Use the preprocessed term as the attribute value, and use the document number as the attribute value address to establish an inverted index; [0005] 3. Search the index table according to the search term input by the user, so as to find the document containing the search term. [0006] There are at least the following problems in the above-mentioned technical scheme: [0007] 1...

AI Technical Summary

Problems solved by technology

The existing technology lists all the documents containing the search term. Although many documents contain the search term, they are less relevant to the search term. Such documents will also be output, resulting in lower accuracy and reliability of the search results.

[0008] 2. Low retrieval efficiency

The existing technology stores all the terms in the document and all the documents containing each term in the index table. Since the number of terms in the document is very large, the number of documents containing each term is also very large, so the index The number of terms in the table and the posting chains of the corresponding documents is quite large, and it takes a lot of resources to search, and the retrieval efficiency is low

Images