Method and device for determining a category directory, and an automatic classification method and device

• US 10,803,091 B2
• Filed: 04/19/2017
• Issued: 10/13/2020
• Est. Priority Date: 10/20/2014
• Status: Active Grant

• Current Assignee: Advanced New Technologies Company Limited (Ant Group Co., Ltd.)
• Original Assignee: Alibaba Group Holding Ltd.
• Inventors: Wei, Peng
• Primary Examiner(s): Pham, Khanh B
• Application Number:
  US15/491,895
• Publication Number:
  US 20170220669A1
• Time in Patent Office: 1,273 Days
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06F 16/285 Clustering or classification

Abstract

Current disclosure discloses a method and device for determining a category directory, and an automatic classification method and device. A user change category message is received. A target change category item is generated based on the user change category message. Previous change category items are analyzed with respect to the target change category item to determine whether an existing category directory needs to be updated.

Claims

1. A device, comprising:

one or more processors;

memory;

a first category directory determination unit, a change category item generation unit, a first time interval determination unit, a first statistical analysis unit, and a second statistical analysis unit that are stored in the memory and executable by the one or more processors, wherein,

the first category directory determination unit is operable to determine a first category directory based on processing side information of resource processing records and a first correspondence between the processing side information and categories, the categories including at least a first category;

the change category item generation unit is operable to receive a change category message from a first user, and to generate a change category item including a change category content indicating changing the first category with a first category ID to a second category with a second category ID;

the first time interval determination unit is operable to determine a statistical starting time associated with the first category ID included in the change category item, and to determine a first time interval associated with the first category ID based on the statistical starting time;

the first statistical analysis unit is operable to statistically analyze change category items that are generated within the first time interval and include the first category ID to obtain a first statistical analysis result; and

the second statistical analysis unit is operable to statistically analyze change category items that are generated within the first time interval in relation to the first user and include a third category ID to obtain a second statistical analysis result, wherein;

the first category directory is set as a default category directory for the first user and one or more other users without changing the first category to the second category if the first statistical analysis result does not meet a first rule and the second statistical analysis result does not meet a second rule, the second rule being different from the first rule,

a second category directory having the first category being changed to the second category is set as the default category directory for the first user and the one or more other users if the first statistical analysis result meets the first rule, and

a user category directory having the first category being changed to the second category is set as a category directory specific for the first user if the first statistical analysis result does not meet a first rule and the second statistical analysis result meets the second rule.

2. The device of claim 1, further comprising a default category directory determination unit and a user category directory determination unit that are stored in the memory and executable by the one or more processors, wherein,

the default category directory determination unit is operable to, if the first statistical analysis result meets the first rule, change the first category directory including the first correspondence to a second category directory including a second correspondence between the processing side information and categories, and set the second category directory as the default category directory, and if the first statistical result does not meet the first rule, set the first category directory as the default category directory;

and

the user category directory determination unit is operable to, if the second statistical analysis result meets the second rule, change the second category directory including the second correspondence to obtain a third category directory including a third correspondence between the processing side information and categories, and set the third category directory as the user category directory for the first user.

3. The device of claim 1, further comprising:

a receiving unit, a default category classification unit and a feedback unit that are stored in the memory and executable by the one or more processors, wherein,

the receiving unit is operable to receive a resource processing record from a second user,

the default category classification unit is operable to classify the resource processing record using the default category directory to obtain a classification result, and

the feedback unit is operable to return the classification result to the second user.

4. The device of claim 3, further comprising a user specific classification unit that is stored in the memory and executable by the one or more processors, wherein the user specific classification unit is operable to:

inquire whether a user category directory associated with the second user from which the resource processing record is received exists, and

if the user category directory associated with the second user exists, classify the received resource processing record based on the user category directory associated with the second user to obtain a classification result,

if no user category directory associated with the second user, transfer to the default category classification unit to classify the received resource processing record.

5. A method comprising:

providing a first category directory including a first correspondence between a processing side information and a first category;

generating a target change category item including a content indicating changing the first category to a second category, upon noticing a first user'"'"'s desire of change;

obtaining a first time interval;

analyzing all previous change category items generated within the first time interval and including the first category to obtain a first analysis result;

analyzing change category items that are generated within the first time interval in relation to the first user and include a third category ID to obtain a second analysis result;

if the first analysis result meets a first rule, modifying the first category directory including the first correspondence to obtain a second category directory including a second correspondence between the processing side information and the second category, and setting the second category directory as a default category directory for the first user and one or more other users;

if the first statistical analysis result does not meet the first rule and the second analysis result meets a second rule, modifying the second category directory including the second correspondence to obtain a third category directory as a user category directory specific for the first user, wherein the second rule is different from the first rule; and

if the first statistical analysis result does not meet the first rule and the second analysis result meets the second rule, setting the first category directory as the default category directory for the first user and the one or more other users.

6. The method of claim 5, wherein the first correspondence between the processing side information and the first category is determined based on analyzing the processing side information, including at least one of a processing party associated with the processing side information, a characteristic of the processing party, and a keyword contained in the processing side information.

7. The method of claim 5, wherein the first correspondence includes one or more processing side information corresponding to the first category.

8. The method of claim 5, wherein the target change category message further includes a user ID, the processing side information, a time of a change category event.

9. The method of claim 5, wherein the previous change category items each include a change category content, a user ID, a time of a change category event, a pre-change category ID and a post-change category ID.

10. The method of claim 5, wherein the analyzing the change category items that are generated within the first time interval and include the first category ID comprises:

identifying a first total amount of first candidate change category items, the first candidate change category items each including a pre-change category ID of the first category;

identifying a second total amount of user ID types associated with the first candidate change category items;

identifying a third total amount of second candidate change category items, the second candidate change category items each including a post-change category ID of the second category; and

identifying a fourth total amount of user ID types associated with the second candidate change category items.

11. The method of claim 10, wherein the first statistical analysis result includes the first total amount, the second total amount, the third total amount and the fourth total amount.

12. The method of claim 10, wherein the first rule comprises at least one of:

the first total amount is larger than or equal to a first threshold, the first threshold being smaller than or equal to a total number of change category items generated within the first time interval;

the second total amount is larger than or equal to a second threshold, the second threshold being smaller than or equal to a total number of user ID types in the change category items generated within the first time interval, and the second threshold being smaller than or equal to the first total amount;

the third total amount is larger than or equal to a third threshold, the third threshold being smaller than or equal to the first total amount; and

the fourth total amount is larger than or equal to a fourth threshold, the fourth threshold being smaller than or equal to the third total amount.

13. The method of claim 5, wherein the analyzing the category items that are generated within the first time interval in relation to the user and include the third category ID comprises:

identifying a fifth total amount of third candidate change category items, the third candidate change category items each including a pre-change category ID same as the third category ID; and

identifying a sixth total amount of fourth candidate change category items, the fourth candidate change category items each being among the third candidate change category items and including a post-change category ID same as a fourth category ID.

14. The method of claim 13, wherein the second analysis result includes the fifth total amount and the sixth total amount.

15. The method of claim 13, wherein the second rule comprises at least one of:

the fifth total amount is larger than or equal to a fifth threshold, the fifth threshold being smaller than or equal to a total number of change category items generated within the first time interval in relation to the first user; and

the sixth total amount in the second analysis result is larger than or equal to a sixth threshold, the sixth threshold being smaller than or equal to the fifth total amount.

16. The method of claim 5, further comprising:

receiving a resource processing record from a second user;

classifying the received resource processing record based on the default category directory; and

returning a result of the classifying to the second user.

17. The method of claim 16, further comprising:

inquiring whether a user category directory associated with the second user exists;

if the user category directory associated with the second user exists, classifying the received resource processing record based the user category directory associated with the second user to obtain a first classification result; and

if no user category directory associated with the second user exists, classifying the received resource processing record based on the default category directory to obtain a second classification result.

18. One or more computer readable media storing executable instructions that, when executed by one or more processors, cause the one or more processors to perform acts comprising:

providing a first category directory including a first correspondence between a processing side information and a first category;

generating a target change category item including a content indicating changing the first category to a second category, upon noticing a desire of change by a user;

analyzing all previous change category items generated within a time interval in relation to the user and including the first category as a pre-change category to obtain a first analysis result;

analyzing change category items that are generated within the time interval in relation to the user and include a third category ID to obtain a second analysis result;

if the first analysis result meets a first rule, modifying the first category directory including the first correspondence to obtain a second category directory including a second correspondence between the processing side information and the second category as a default category directory for the first user and one or more other users;

if the first statistical analysis result does not meet the first rule and the second analysis result meets a second rule, modifying the second category directory including the second correspondence to obtain a third category directory as a user category directory specific for the user, wherein the second rule is different from the first rule; and

if the first statistical analysis result does not meet the first rule and the second analysis result meets the second rule, setting the first category directory as the default category directory for the user and the one or more other users.

19. The one or more computer readable media of claim 18, wherein the analyzing all previous change category items generated within the time interval in relation to the user and including the first category as a pre-change category includes:

obtaining a fifth total amount of a first candidate change category items, the first candidate change category items each including a pre-change category ID of the first category; and

obtaining a sixth total amount of second candidate change category items, the second candidate change category items each being among the first candidate change category items and including a post-change category ID of the second category.