Systems, methods, and computer program products for expediting expertise

Patent metadata

Field	Value
Publication number	US-9275332-B2
Application number	US-201213648988-A
Country	US
Kind code	B2
Filing date	Oct 10, 2012
Priority date	Oct 10, 2012
Publication date	Mar 1, 2016
Grant date	Mar 1, 2016

Abstract

A method including generating a global topic model based on a set of data that is updated according to an activity of each user of a plurality of users, the global topic model including a topic representation for a topic, generating a plurality of user models, each user model being generated based on the activity of a respective user, generating an expertise model for the topic based on the activity of at least one user of the plurality of users, the expertise model for the topic setting a target level of knowledge for a first user of the plurality of users, comparing a user model of the first user with the expertise model for the topic, the comparing being performed by a processor of a computer system, and recommending an activity associated with the set of data to the first user based on the comparison.

First claim

What is claimed is:

1. A method, comprising:
generating, as executed by a processor on a computer, a plurality of topic-specific user knowledge models for each user of a plurality of users, each topic-specific user knowledge model representing a level of knowledge possessed by a respective user on a single topic from a set of globally defined topics shared among the plurality of users;
generating a plurality of topic-specific expert knowledge models, each topic-specific expert knowledge model representing an aggregate level of knowledge possessed by a plurality of expert users on a single topic from a set of globally defined topics shared among the plurality of users;
comparing the topic-specific user knowledge model of the first user with the topic-specific expert knowledge model for a respective topic to determine a distance between a user knowledge level and an aggregate expert knowledge level for the topic, the comparing being performed by the processor on the computer;
recommending an activity including at least one of reading, writing, downloading, watching, and commenting of a content item relevant to the topic to the first user based on said comparing;
updating the topic-specific user knowledge model of the first user to reflect the first user's increased level of knowledge after the first user engages in the recommended activity for the topic; and
adjusting the activity recommendation based on the updated topic-specific user knowledge model.

2. The method according to claim 1 , wherein, in the recommending, the activity includes said reading, said writing, said downloading, said commenting, and said watching.

3. The method according to claim 1 , wherein said each of the topic-specific user knowledge models comprises a topic-specific user activity vector, where each dimension of the vector represents the user's knowledge level as reflected in a certain type of a user's activities including said reading, said writing, said downloading, said watching, and said commenting of the respective user on the respective topic.

4. The method according to claim 3 , wherein calculation of the topic-specific user activity vector comprises:
combining a topic relevance, a popularity, and an importance of a content item associated with a single user activity of the respective user to calculate a topic-specific score of the activity.

5. The method according to claim 4 , wherein said calculation of the topic-specific user activity vector further comprises:
summing up the topic-specific scores of all user activities of the respective user that are of a certain activity type to determine a topic- and type-specific score of a respective activity type for the respective user.

6. The method according to claim 5 , wherein said calculation of the topic-specific user activity vector further comprises:
applying a type-specific weight to the topic- and type-specific score of the respective activity type to determine a value of a respective dimension in the user activity vector.

7. The method according to claim 1 , wherein the topic-specific expert knowledge model comprises a topic-specific expert activity vector.

8. The method according to claim 7 , wherein calculation of the expert activity vector comprises:
selecting user activity vectors of the plurality of users who are determined as experts of the respective topic.

9. The method according to claim 8 , wherein said calculation of the expert activity vector further comprises:
aggregating values for a certain dimension of the selected user activity vectors to calculate a value of a respective dimension for an expert activity vector.

10. The method according to claim 9 , wherein said aggregating comprises calculating summation, maximum, minimum, and mean values for the certain dimension.

11. The method according to claim 1 , wherein the distance between the user knowledge level and the aggregate expert knowledge level for the single topic is calculated based on a vector distance of a topic-specific user activity vector and a topic-specific expert activity vector.

12. The method according to claim 1 , wherein the recommended activity includes a certain type of activity to be performed on a content associated with activities of the expert users.

13. The method according to claim 12 , wherein a ranking of the recommended activity is determined based on an expected return-on-investment value whose calculation comprises:
an expected change in a user's current knowledge level if the user engages in the recommended activity;
a popularity and an importance of an associated content;
a similarity or a dissimilarity of the recommended activity with existing activities of the user; and
an estimated cost of performing the recommended activity.

14. The method according to claim 1 , wherein the updating of the topic-specific user knowledge model of the first user is dynamically performed after the first user engages in a recommended activity for the topic.

15. The method according to claim 14 , wherein the updating comprises:
combining a topic relevance, a popularity, and an importance of a content item associated with an accomplished recommended activity of the respective user to calculate a topic-specific score of the activity;
applying a type-specific weight a the topic-specific score of an activity to determine a topic- and type-specific score of the activity; and
adding the topic- and type-specific score of the activity to a respective dimension in a topic-specific user activity vector based on a type of the accomplished recommended activity.

Assignees

• IBM

Inventors

• Lu Jie
• Boston Jeffrey Scott
• Coden Anni Rosa
• Lai Jennifer Ceil
• Pan Shimei
• Topkara Mercan
• Wen Zhen
• Wood Stephen Peter

Classifications

• G06N5/02 Primary

  Knowledge representation; Symbolic representation · CPC title

• G06Q10/06

  Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling · CPC title

• G06N5/04 Primary

  Inference or reasoning models · CPC title