US Patent Application 18446034. Information Retrieval Method, Related System, and Storage Medium simplified abstract
Contents
Information Retrieval Method, Related System, and Storage Medium
Organization Name
Huawei Technologies Co., Ltd.==Inventor(s)==
[[Category:Xinyu Zhang of Hangzhou (CN)]]
[[Category:Ke Zhan of Shenzhen (CN)]]
[[Category:Lan Luo of Hangzhou (CN)]]
[[Category:Chengzhen Fu of Shenzhen (CN)]]
[[Category:Enrui Hu of Hangzhou (CN)]]
Information Retrieval Method, Related System, and Storage Medium - A simplified explanation of the abstract
This abstract first appeared for US patent application 18446034 titled 'Information Retrieval Method, Related System, and Storage Medium
Simplified Explanation
The abstract describes an information retrieval method that involves obtaining candidate documents based on a retrieval text query and scoring them. It also involves calculating probabilities of paths corresponding to the candidate documents and obtaining additional candidate documents based on these probabilities. Finally, it obtains a retrieval result corresponding to the query based on the candidate documents.
- The method is used for information retrieval based on a text query.
- It involves obtaining candidate documents at different hop levels.
- The candidate documents are scored to determine their relevance.
- Probabilities of paths corresponding to the candidate documents are calculated.
- Additional candidate documents are obtained based on these probabilities.
- The final retrieval result is obtained based on the candidate documents.
Original Abstract Submitted
An information retrieval method includes obtaining Mi (i+1)-hop candidate documents based on a retrieval text query and Ki i-hop candidate documents; obtaining a score of each candidate document in the Mi (i+1)-hop candidate documents; obtaining, based on a score of a candidate document Pjy(i+1) and a probability of a path L, a probability of a path corresponding to the candidate document Pjy(i+1); obtaining K(i+1) (i+1)-hop candidate documents based on probabilities of paths respectively corresponding to the Mi (i+1)-hop candidate documents; and obtaining, based on the K(i+1) (i+1)-hop candidate documents, a retrieval result corresponding to the query.
