Discovering communities and expertise of users using semantic analysis of resource access logs

• US 9,852,208 B2
• Filed: 02/25/2014
• Issued: 12/26/2017
• Est. Priority Date: 02/25/2014
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Chari, Suresh N., Molloy, Ian M., Park, Youngja
• Primary Examiner(s): Le, Miranda
• Application Number:
  US14/189,274
• Publication Number:
  US 20150242486A1
• Time in Patent Office: 1,400 Days
• Field of Search: 707737
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: H04L 63/104 Grouping of entities
• CPC Class Codes: H04L 65/403 Arrangements for multi-party communication, e.g. for conferences data switching systems for conference H04L12/18; arrangements for connecting several subscribers to a common circuit, i.e. affording conference facilities H04M3/56; television conferencing systems H04N7/15
• CPC Class Codes: H04L 67/306 User profiles
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

Generating communities of users and discovering the expertise of those users are provided. Identifications of a plurality of users that accessed resources via a network, types of actions performed by the plurality of users on the resources, and names of the resources accessed by the plurality of users are extracted from retrieved resource access logs. The plurality of users are grouped into a plurality of different sets of users based on which resources were accessed and which type of actions were performed by each particular user within the plurality of users. The communities of users are generated based on each different set of users having a similarity with regard to the resources that were accessed and the types of actions that were performed by a particular set of users.

Claims

1. A computer-implemented method for generating communities of users, the computer-implemented method comprising:

extracting, by a computer, identifications of a plurality of users that accessed resources via a network, types of actions performed by the plurality of users on the resources, and names of the resources accessed by the plurality of users from retrieved resource access logs;

grouping, by the computer, the plurality of users into a plurality of different sets of users based on which resources were accessed and which type of actions were performed by each particular user within the plurality of users;

generating, by the computer, the communities of users based on each different set of users having a similarity with regard to the resources that were accessed and the types of actions that were performed by a particular set of users;

generating, by the computer, a user matrix for the plurality of users that accessed the resources using the types of actions performed by the plurality of users on the resources and the names of the resources accessed by the plurality of users extracted from the retrieved resource access logs;

generating, by the computer, a row in the user matrix for a particular user as a feature vector for the particular user, wherein the feature vector includes all action/resource pairs associated with the particular user within the plurality of users;

generating, by the computer, the feature value within the feature vector for an action/resource pair associated with the particular user equal to the number of access logs that record the particular user performing the particular action on the particular resource and to the log of the number of all user/action/resource groupings within the access logs divided by the number of unique users that performed the particular action on the particular resource;

generating, by the computer, using information extracted from a retrieved resource access log, a user/action/resource grouping based on an identifier of a particular user, a type of action performed by the particular user on a particular resource, and the name of the particular resource accessed by the particular user;

retrieving, by the computer, the type of action performed by the particular user on the particular resource from the user/action/resource grouping; and

displaying, by the computer, the user matrix with only those action/resource pairs that do not exceed a threshold for the feature value.

2. The computer-implemented method of claim 1, further comprising:

labeling, by the computer, each community of users with labels that represent one of roles or expertise of users assigned to a particular community.

3. The computer-implemented method of claim 2, wherein the computer determines at least one of most relevant resource names and action types for each community of users to label different communities of users.

4. The computer-implemented method of claim 1, further comprising:

responsive to the computer determining that the name of the particular resource within the user/action/resource grouping is expressed in natural language, segmenting, by the computer, the name of the particular resource at, at least one of, each punctuation mark, each file name separator, and each upper cased letter contained within the name of the particular resource to generate a set of resource name segments that corresponds to the particular resource

generating, by the computer, an action/resource pair for each resource name segment within the set of resource name segments by combining the retrieved type of action performed by the particular user on the particular resource with each of the resource name segments.

5. The computer-implemented method of claim 1, further comprising:

generating, by the computer, a feature set for the plurality of users from retrieved resource access logs corresponding to all pairs of types of actions performed by the plurality of users and names of resources accessed by the plurality of users by performing a particular action.

6. The computer-implemented method of claim 1, wherein the computer utilizes one of a clustering algorithm or a topic modeling algorithm over the user matrix to determine roles, expertise, and attributes of users associated with a particular community of users, and wherein the computer generates a model of the particular community of users based on the roles, expertise, and attributes of the users associated with the particular community of users.

7. The computer-implemented method of claim 6, wherein the computer assigns users to a community based on the attributes of the users.

8. The computer-implemented method of claim 1, wherein the computer defines a role that provides members of a particular community of users with permissions to access resources associated with that particular community of users.

9. The computer-implemented method of claim 1, wherein the computer assigns a set of resources to a particular community of users based on members of that particular community of users previously accessing the set of resources.

10. The computer-implemented method of claim 1, wherein the computer assigns a set of resources to a particular community of users based on resource name patterns specified by members of that particular community of users during prior accesses to the set of resources.

11. The computer-implemented method of claim 10, wherein the computer assigns the set of resources to the particular community of users based on order of names within the resource name patterns.

12. The computer-implemented method of claim 1, wherein the computer assigns a role derived from a community of users to all resources in sub-branches of a particular file path, wherein all resources in the sub-branches are assigned to the community having a likelihood of being included in a particular group of resources equal to or greater than a predetermined threshold value.

13. The computer-implemented method of claim 1, wherein the computer assigns a sensitivity level to a particular community of users based on labels associated with the particular community of users.

14. The computer-implemented method of claim 1, wherein the computer sends an alert message in response to a particular user transitioning from a lower sensitivity level community of users to a higher sensitivity level community of users that the particular user did not previously participate in.

15. A computer system for generating communities of users, the computer system comprising:

a bus system;

a storage device connected to the bus system, wherein the storage device stores computer readable program code; and

a processor unit connected to the bus system, wherein the processor unit executes the computer readable program code to extract identifications of a plurality of users that accessed resources via a network, types of actions performed by the plurality of users on the resources, and names of the resources accessed by the plurality of users from retrieved resource access logs;

group the plurality of users into a plurality of different sets of users based on which resources were accessed and which type of actions were performed by each particular user within the plurality of users; and

generate the communities of users based on each different set of users having a similarity with regard to the resources that were accessed and the types of actions that were performed by a particular set of users;

the processor unit further executes computer readable code to generate a user matrix for the plurality of users that accessed the resources using the types of actions performed by the plurality of users on the resources and the names of the resources accessed by the plurality of users extracted from the retrieved resource access logs;

to generate a row in the user matrix for a particular user as a feature vector for the particular user, wherein the feature vector includes all action/resource pairs associated with the particular user within the plurality of users;

to generate a feature value within the feature vector for an action/resource pair associated with the particular user equal to the number of access logs that record the particular user performing the particular action on the particular resource and to the log of the number of all user/action/resource groupings within the access logs divided by the number of unique users that performed the particular action on the particular resource;

to generate, using information extracted from a retrieved resource access log, a user/action/resource grouping based on an identifier of a particular user, a type of action performed by the particular user on a particular resource, and the name of the particular resource accessed by the particular user;

to retrieve the type of action performed by the particular user on the particular resource from the user/action/resource grouping; and

to display the user matrix with only those action/resource pairs that do not exceed a threshold for the feature value.

16. A computer program product stored on a non-transitory computer readable storage medium having computer readable program code encoded thereon that is executable by a computer for generating communities of users, the computer program product comprising:

computer readable program code to extract identifications of a plurality of users that accessed resources via a network, types of actions performed by the plurality of users on the resources, and names of the resources accessed by the plurality of users from retrieved resource access logs;

computer readable program code to group the plurality of users into a plurality of different sets of users based on which resources were accessed and which type of actions were performed by each particular user within the plurality of users;

computer readable program code to generate the communities of users based on each different set of users having a similarity with regard to the resources that were accessed and the types of actions that were performed by a particular set of users;

computer readable program code to generate a user matrix for the plurality of users that accessed the resources using the types of actions performed by the plurality of users on the resources and the names of the resources accessed by the plurality of users extracted from the retrieved resource access logs;

computer readable program code to generate a row in the user matrix for a particular user as a feature vector for the particular user, wherein the feature vector includes all action/resource pairs associated with the particular user within the plurality of users;

computer readable program code to generate a feature value within the feature vector for an action/resource pair associated with the particular user equal to the number of access logs that record the particular user performing the particular action on the particular resource and to the log of the number of all user/action/resource groupings within the access logs divided by the number of unique users that performed the particular action on the particular resource;

computer readable program code to generate using information extracted from a retrieved resource access log, a user/action/resource grouping based on an identifier of a particular user, a type of action performed by the particular user on a particular resource, and the name of the particular resource accessed by the particular user;

computer readable program code to retrieve the type of action performed by the particular user on the particular resource from the user/action/resource grouping; and

computer readable program code to display the user matrix with only those action/resource pairs that do not exceed a threshold for the feature value.