Comment ranking by search engine
AssigneeAlibaba Group Holding Limited
InventorsRanran Du, Yongbin Lin
Life of this patent
Estimating the term…
Abstract
The present disclosure provides example methods and apparatuses for implementing comment ranking by a search engine. One or more terms describing a target object are extracted from a comment in connection with the target object. One or more useful attributes included in the comment are obtained according to the terms describing the target object. An scoring index the comment is determined according to a number of the useful attributes. The comment is scored according to the index. The comment is ranked according to the score. The present techniques enable a user to quickly and efficiently view helpful comments and help the user make an informed decision.
Claims20
A method, the method performed by a server, the server comprising one or more processors and a memory device communicatively coupled with the one or more processors, the memory device including instructions, which when executed cause the one or processors to perform acts comprising:
- extracting one or more terms describing a target object from a comment with respect to the target object;
- identifying one or more emotional terms in the one or more terms describing the target from a list of emotional terms maintained at the server;
- determining one or more scoring indexes to score the comment according to a number of useful attributes in the comment based on the one or more emotional terms, the one or more scoring indexes including readability of the comment and a ranking factor of one or more useless attributes in the comments, the readability of the comment determined based on a normalized length of clauses that is calculated based on associated lengths of all of clauses in the comment;
- scoring the comment according to the one or more scoring indexes;
- and
- ranking the comment according to a result of the scoring.
An apparatus comprising:
- one or more processors;
- and
- a memory device including instructions executable by the one or more processors which when executed perform acts comprising:
- extracting one or more terms describing a target object from a comment with respect to the target object;
- identifying one or more emotional terms in the one or more terms describing the target from a list of emotional terms maintained at the server;
- obtaining a number of useful attributes in the comment based on the one or more emotional terms;
- determining one or more scoring indexes to score the comment according to a number of useful attributes in the comment based on the one or more emotional terms, the one or more scoring indexes including readability of the comment and a ranking factor of one or more useless attributes in the comments, the readability of the comment determined based on a normalized length of clauses that is calculated based on associated lengths of all of clauses in the comment;
- scoring the comment according to the one or more scoring indexes;
- and
- ranking the comment according to a result of the scoring.
One or more computer storage media storing thereon computer-executable instructions executable by one or more processors to perform operations comprising:
- extracting one or more terms describing a target object from a comment with respect to the target object;
- identifying one or more emotional terms in the one or more terms describing the target from a list of emotional terms maintained at the server;
- determining one or more scoring indexes to score the comment according to a number of useful attributes in the comment based on the one or more emotional terms, the one or more scoring indexes including readability of the comment and a ranking factor of one or more useless attributes in the comments, the readability of the comment determined based on a normalized length of clauses that is calculated based on associated lengths of all of clauses in the comment;
- scoring the comment according to the one or more scoring indexes;
- and
- ranking the comment according to a result of the scoring.
Dependencies are inferred from the claim text.