Search error correction method, apparatus and device, computer program product

Abstract

This application discloses a search error correction method, apparatus, device, and computer program product. The method includes: acquiring and preprocessing query information input by a user; determining the corresponding error correction link based on the type of the preprocessed query information; if the corresponding error correction link is a Chinese character error correction link, performing error correction processing on the preprocessed query information using a multi-level Chinese character error correction strategy to obtain a Chinese character search error correction result; if the corresponding error correction link is a Pinyin error correction link, performing error correction processing on the preprocessed query information using a Pinyin error correction strategy to obtain a Pinyin search error correction result. This application improves the targeting, accuracy, and efficiency of search error correction by distinguishing the type of query information and routing it to different error correction links, and by employing independent and appropriate processing strategies and dictionary systems for Chinese characters and Pinyin respectively, providing users with a better search experience and reducing search failures caused by input errors.

Claims

1. A search error correction method, characterized in that, The search error correction method includes: Obtain the query information input by the user and preprocess it to obtain the preprocessed query information; The type of the preprocessed query information is determined, and the error correction link corresponding to the preprocessed query information is determined according to the type of the preprocessed query information. The error correction link is divided into Chinese character error correction link and Pinyin error correction link. When the error correction link corresponding to the preprocessed query information is the Chinese character error correction link, the preprocessed query information is corrected using a multi-level Chinese character error correction strategy to obtain the Chinese character search error correction result. If the error correction link corresponding to the preprocessed query information is the Pinyin error correction link, the Pinyin error correction strategy is used to perform error correction processing on the preprocessed query information to obtain the Pinyin search error correction result.

2. The search error correction method according to claim 1, characterized in that, The step of using a multi-level Chinese character error correction strategy to correct errors in the preprocessed query information to obtain Chinese character search error correction results includes: Based on the preprocessed query information, a search is performed in the pre-constructed correct vocabulary list to obtain the first search result; If the first search result finds a correct word that matches the preprocessed query information, the correct word that matches the preprocessed query information is taken as the Chinese character search error correction result.

3. The search error correction method according to claim 2, characterized in that, The step of using a multi-level Chinese character error correction strategy to correct errors in the preprocessed query information to obtain Chinese character search error correction results includes: If the first search result does not find a correct word that matches the preprocessed query information, then a search is performed in the pre-built error-correct word mapping table to obtain a second search result; If the second search result finds a correct word that matches the preprocessed query information, the correct word that matches the preprocessed query information is taken as the Chinese character search error correction result.

4. The search error correction method according to claim 3, characterized in that, The step of using a multi-level Chinese character error correction strategy to correct errors in the preprocessed query information to obtain Chinese character search error correction results includes: If the second search result does not find the correct word that matches the preprocessed query information, the search is sorted according to the preprocessed query information using a preset search sorting strategy to obtain the search sorting result. The preset search sorting strategy includes a search strategy based on BK tree structure and a sorting strategy based on multi-dimensional features and a predefined bounded priority queue. The Chinese character search error correction result is determined based on the search ranking result.

5. The search error correction method according to claim 4, characterized in that, The BK tree structure is constructed based on a correct vocabulary. The search ranking is performed using a preset search ranking strategy based on the preprocessed query information, and the resulting search ranking includes: The preprocessed query information is searched using the BK tree-based search strategy to obtain candidate words corresponding to the preprocessed query information; The candidate words corresponding to the query information are sorted using the multi-dimensional feature-based sorting strategy and a predefined bounded priority queue to obtain the sorting result of the candidate words. The multi-dimensional features include word frequency, pinyin similarity, character structure, edit distance, and word length difference.

6. The search error correction method according to claim 1, characterized in that, The step of using a pinyin error correction strategy to correct errors in the preprocessed query information to obtain pinyin search error correction results includes: Based on the preprocessed query information, a search is performed in the pre-constructed Pinyin-Chinese character mapping dictionary to obtain the third search result; The pinyin search error correction result is determined based on the third search result.

7. The search error correction method according to any one of claims 3 to 5, characterized in that, The error-correct vocabulary mapping table is constructed as follows: A Chinese character confusion relation database is constructed from multiple dimensions, including at least one of the following: Chinese character structural similarity, Chinese character pronunciation features, and Chinese character stroke and structure database. Based on the Chinese character confusion relation database, the error-correct word mapping table is automatically generated using a rule-driven approach.

8. The search error correction method according to claim 7, characterized in that, The search error correction method also includes: The error-correct vocabulary mapping table is updated using a multi-dimensional update strategy, which includes at least one of the following: an existing product name database, a function menu list, and a search engine thesaurus.

9. A search error correction device, characterized in that, The search and error correction device includes: The acquisition unit is used to acquire the query information input by the user and preprocess it to obtain the preprocessed query information; The determining unit is used to determine the type of the preprocessed query information and to determine the error correction link corresponding to the preprocessed query information based on the type of the preprocessed query information. The error correction link is divided into Chinese character error correction link and Pinyin error correction link. The first search error correction unit is used to perform error correction processing on the preprocessed query information using a multi-level Chinese character error correction strategy when the error correction link corresponding to the preprocessed query information is the Chinese character error correction link, so as to obtain the Chinese character search error correction result. The second search error correction unit is used to perform error correction processing on the preprocessed query information using a pinyin error correction strategy when the error correction link corresponding to the preprocessed query information is the pinyin error correction link, so as to obtain the pinyin search error correction result.

10. An apparatus comprising: processor; And a memory arranged to store computer-executable instructions, which, when executed, cause the processor to perform the search and error correction method according to any one of claims 1 to 8.

11. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the search and error correction method according to any one of claims 1 to 8.

Images