METHOD OF DATA QUERY BASED ON EVALUATION AND DEVICE

• US 20190332602A1
• Filed: 07/01/2019
• Published: 10/31/2019
• Est. Priority Date: 07/02/2018
• Status: Active Grant

• Current Assignee: Beijing Baidu Netcom Science & Technology Co. Ltd.
• Original Assignee: Beijing Baidu Netcom Science & Technology Co. Ltd.
• Inventors: XU, Weijian
• Granted Patent: US 11,176,142 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/215 Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
• CPC Class Codes: G06F 16/242 Query formulation
• CPC Class Codes: G06F 16/24575 using context
• CPC Class Codes: G06F 16/9035 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/953 Querying, e.g. by the use of web search engines
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0631 Item recommendations

Abstract

A method of data query based on an evaluation and a device. The method includes: obtaining evaluation data of entities and basic information of the entities from multiple websites; extracting labels of the entities according to the evaluation data of the entities and the basic information of the entities; filtering the evaluation data of the entities; aggregating to obtain evaluation data of the same type of entities having the same label according to the basic information of the entities, the labels of the entities and the filtered evaluation data of the entities; and making a query to obtain information on a target entity according to a retrieval statement and the aggregated evaluation data of the same type of entities for each label. The method can make a query for information on a type of entities for a user, so that the user can compare the entities.

Claims

1. A method of data query based on an evaluation, comprising:

obtaining evaluation data of entities and basic information of the entities from multiple websites;

extracting labels of the entities according to the evaluation data of the entities and the basic information of the entities, wherein the labels are used to indicate user'"'"'s views on the entities;

filtering the evaluation data of the entities;

aggregating to obtain evaluation data of the same type of entities having the same label according to the basic information of the entities, the labels of the entities and the filtered evaluation data of the entities; and

making a query to obtain information on a target entity according to a retrieval statement and the aggregated evaluation data of the same type of entities for each of the labels, wherein the retrieval statement comprises a type and a label of an entity, the target entity is a type of entities, each of which matches the type and the label of the entity comprised in the retrieval statement, and the information on the target entity comprises basic information of the target entity and evaluation information of the target entity.

2. The method according to claim 1, before the extracting of labels of the entities according to the evaluation data of the entities and the basic information of the entities, further comprising:

obtaining a retrieval term and retrieval times of the retrieval term corresponding to each type of entities in a preset period;

the extracting labels of the entities according to the evaluation data of the entities and the basic information of the entities comprises;

extracting the labels of the entities according to the evaluation data of the entities, the basic information of the entities, and the retrieval term and the retrieval times of the retrieval term corresponding to each type of entities in the preset period.

3. The method according to claim 2, wherein the obtaining of basic information of the entities from multiple websites comprises:

obtaining original information of the entities from the multiple websites, wherein the original information comprises a name, a type, and a keyword of an entity; and

identifying the same entity according to names, types and keywords of the entities, and obtaining the basic information of the entities.

4. The method according to claim 1, wherein the filtering of the evaluation data of the entities comprises:

performing quality scoring on evaluations of the entities, and deleting evaluation data whose quality scoring is lower than a score threshold; and

performing sensitive word marking on the evaluations of the entities, and deleting evaluation data containing a sensitive word.

5. The method according to claim 4, before the aggregating to obtain evaluation data of the same type of entities having the same label according to the basic information of the entities, the labels of the entities and the filtered evaluation data of the entities, further comprising:

performing matching on the labels of the entities, and deleting a label of an entity that does not match the type of the entity from the labels of the entities; and

performing emotion analysis on the evaluations of the entities to obtain emotional scores of the entities.

6. The method according to claim 5, further comprising:

obtaining emotional fluctuation data of the entities according to the emotional scores of the entities; and

the information on the target entity further comprises the emotional fluctuation data of the entity.

7. The method according to claim 1, after the aggregating to obtain evaluation data of the same type of entities having the same label, further comprising:

performing deduplication on the evaluation data of the same type of entities for each of the labels.

15. A server, comprising a processor, a memory and a transceiver, wherein the memory is configured to store instructions, the transceiver is configured to communicate with other device, the processor is configured to execute the instructions stored in the memory, so that the server performs the method according to claim 1.

16. The method according to claim 1, further comprising:

executing the method on a computer readable medium that stores instructions.

8. A device of data query based on an evaluation, comprising:

a processor and a computer-readable medium for storing program codes, which, when executed by the processor, cause the processor to;

obtain evaluation data of entities and basic information of the entities from multiple websites;

extract labels of the entities according to the evaluation data of the entities and the basic information of the entities, wherein a label is used to indicate a user'"'"'s view on an entity;

filter the evaluation data of the entities;

aggregate to obtain evaluation data of the same type of entities having the same label according to the basic information of the entities, the labels of the entities, and the filtered evaluation data of the entities; and

make a query to obtain information on a target entity according to a retrieval statement and the aggregated evaluation data of the same type of entities for each of the labels, wherein the retrieval statement comprises a type and a label of an entity, the target entity is a type of entities, each of which matches the type and the label of the entity comprised in the retrieval statement, and the information on the target entity comprises basic information of the target entity and evaluation information of the target entity.

9. The device according to claim 8, wherein the program codes further cause the processor to:

obtain a retrieval term and retrieval times of the retrieval term corresponding to each type of entities in a preset period; and

extract the labels of the entities according to the evaluation data of the entities, the basic information of the entities, and the retrieval term and the retrieval times of the retrieval term corresponding to each type of entities in the preset period.

10. The device according to claim 9, wherein the program codes further cause the processor to:

obtain original information of the entities from the multiple websites, wherein the original information comprises a name, a type, and a keyword of an entity; and

identify the same entity according to names, types and keywords of the entities, to obtain basic information of the entities.

11. The device according to claim 8, wherein the program codes further cause the processor to:

perform quality scoring on evaluations of the entities, and delete evaluation data whose quality scoring is lower than a score threshold; and

perform sensitive word marking on the evaluations of the entities, and delete evaluation data containing a sensitive word.

12. The device according to claim 11, wherein the program codes further cause the processor to:

perform matching on the labels of the entities, and delete a label of an entity that does not match the type of the entity from the labels of the entities;

perform emotion analysis on the evaluations of the entities, to obtain emotional scores of the entities; and

sort the evaluations of the entities according to an access popularity, or to scatter and sort the evaluations of the entities.

13. The device according to claim 12, wherein the program codes further cause the processor to:

obtain emotional fluctuation data of the entities according to the emotional scores of the entities; and

the information on the target entity further comprises the emotional fluctuation data of the entity.

14. The device according to claim 8, wherein the program codes further cause the processor to:

perform deduplication on the evaluation data of the same type of entities for each of the labels.