Search method and system for tag recognition of no-result words
A technology of label identification and search method, applied in the field of Internet e-commerce, can solve the problems of increasing search secondary recall, search word digestion, no search result return, etc., to achieve the effect of expanding the scope of crawling
Active Publication Date: 2019-08-16
SUNING COM CO LTD
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0002] At present, search technology only has modules such as word segmentation, synonym processing, error correction, and word reduction for the processing of search terms. Many search words with high user attention cannot be digested by error correction or word reduction, that is, after the keyword is processed by error correction and word reduction, no search results are finally returned
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0040] The present invention will be further described below in conjunction with embodiment and accompanying drawing.
[0041] There are two kinds of existing processing methods for no result words: error correction and word reduction. Among them, error correction mainly takes no result words as the core. The similarity score before the word is obtained, and the similar word with a high similarity with the original word is obtained as a replacement word. Word reduction refers to the selection of the result of the word segmentation process without the result, and the part with the result is reserved as the replacement word of the original word.
[0042] The system module data flow of the inventive method is as figure 1 As shown, firstly crawl the external network data for the search words that still have no results after error correction processing and word reduction search (that is, no result words), segment the crawled documents and extract high-frequency words , where the ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a searching method and system for carrying out label identification on a resultless word. According to the searching method and system, starting from a user search word, a real search intention of a user is analyzed, related data of an Extranet is crawled, then a crawling result is subjected to data mining to extract out keywords with high relevance degrees with the word, i.e., high frequency words, and simultaneously, common corpora are combined to carry out classification on the high frequency words. According to the searching method and system disclosed by the invention, a crawling range of the related data is enlarged, and the defect that a common resultless processing method is mainly limited to mining on search data and user behaviors in station and is easy to generate a processing failure is overcome.
Description
technical field [0001] The invention belongs to the field of Internet e-commerce, and relates to a secondary search scene when no result is returned in a commodity search in the Internet e-commerce, in particular to a search method and system for performing label recognition on no result words and generating replacement search words. Background technique [0002] At present, search technology only has modules such as word segmentation, synonym processing, error correction, and word reduction for the processing of search terms. Most of the search terms with high user attention cannot be digested by error correction or word reduction, that is, after the keyword is processed by error correction and word reduction, there is still no search result returned in the end. [0003] Search terms with high user attention (such as search terms with more than 20 daily searches) are mainly inoperable commodity words that include brand, category, and attribute information. The inoperable co...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/33G06F16/951G06F17/27
Inventor 贾洪园张晶
Owner SUNING COM CO LTD