Estimating data topics of computers using external text content and usage information of the users

• US 9,600,577 B2
• Filed: 09/26/2013
• Issued: 03/21/2017
• Est. Priority Date: 08/01/2013
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Gates, Christopher, Park, Youngja
• Primary Examiner(s): Uddin, Mohammed R
• Application Number:
  US14/038,220
• Publication Number:
  US 20150039618A1
• Time in Patent Office: 1,272 Days
• Field of Search: 707/738, 707/739, 707/748, 707/750, 707/999.009
• US Class Current: 1/1
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

Automatically estimating content topics of inaccessible content in a computer system, in one aspect, may comprise gathering accessible content and analyzing the accessible content to estimate one or more topics of the inaccessible content.

Claims

1. A system to automatically estimate content topics of inaccessible content in a computer system without inspecting data files in the computer system comprising:

a processor; and

a module operator to execute on the processor and further operable to gather accessible content, the module further operator to analyze the accessible content to estimate one or more topics of the inaccessible content without inspecting the inaccessible content,

the inaccessible content comprising privileged data protected from access due to one or more of data privacy and computer security, wherein the one or more topics of the inaccessible content is estimated while preserving the one or more of data privacy and computer security,

the module estimating the one or more topics at least by;

identifying users of the computer system and access counts of the users accessing the computer system, retrieving the accessible content generated by the users of the computer system, analyzing user information and external text content associated with the users that are available in an organization'"'"'s online space outside of the computer system;

for each of the users, generating a document comprising a bag-of-words representation for the inaccessible content generated by the user, the bag-of-words representation comprising words occurring in the accessible content and counts of the words, the counts of the words scaled as a function of a number of occurrences of a word in the accessible content and a computer system access count associated with the user;

generating an asset document associated with the computer system by aggregating the document associated with each user for all users; and

executing a topic modeling algorithm on the asset document that estimates the one or more topics,

wherein based on the one or more topics, the module automatically determines security level of information stored in the computer system.

2. The system of claim 1, wherein the accessible content comprises accessible content at least created or consumed by a user.

3. The system of claim 1, wherein the accessible content is distributed over a set of enterprise machines.

4. The system of claim 1, wherein the module is further operable to apply a generative topic modeling algorithm to analyze the accessible content.

13. The system of claim 1, wherein the accessible content and the inaccessible content are separately stored content.

5. A computer readable storage device storing a program of instructions executable by a machine to perform a method to automatically estimate content topics of inaccessible content in a computer system, the method comprising:

gathering accessible content; and

analyzing the accessible content to estimate one or more topics of the inaccessible content without inspecting the inaccessible content,

the inaccessible content comprising privileged data protected from access due to one or more of data privacy and computer security, wherein the one or more topics of the inaccessible content is estimated while preserving the one or more of data privacy and computer security,

the analyzing further comprising;

identifying users of the computer system and access counts of the users accessing the computer system, retrieving the accessible content generated by the users of the computer system outside of the computer system, analyzing user information and external text content associated with the users that are available in an organization'"'"'s online space outside of the computer system;

for each of the users, generating a document comprising a bag-of-words representation for the inaccessible content generated by the user, the bag-of-words representation comprising words occurring in the accessible content and counts of the words, the counts of the words scaled as a function of a number of occurrences of a word in the accessible content and a computer system access count associated with the user;

generating an asset document associated with the computer system by aggregating the document associated with each user for all users; and

executing a topic modeling algorithm on the asset document that estimates the one or more topics,

wherein based on the one or more topics, the module automatically determines security level of information stored in the computer system.

6. The computer readable storage device of claim 5, wherein the accessible content comprises accessible content at least created or consumed by a user.

7. The computer readable storage device of claim 6, wherein the accessible content created or consumed by a user comprises at least one of website content, publication, patent, employee record, file sharing repository, bookmarks, and tags.

8. The computer readable storage device of claim 5, wherein the gathering accessible content comprises:

identifying one or more users that accessed the computer system;

for each of the one or more users,

collecting content that are accessible and that said each user has at least created or consumed, and

creating a pseudo document associated with said each user; and

combining the collected content of all of the one or more users to create a document associated with the inaccessible content.

9. The computer readable storage device of claim 8, further comprising:

for each of the one or more users, determining a frequency of access to the computer system by said each user, and scoring the pseudo document based on the frequency.

10. The computer readable storage device of claim 8, wherein the analyzing the accessible content comprises applying a topic modeling algorithm on the document associated with the inaccessible content.

11. The computer readable storage device of claim 10, wherein the algorithm comprises Latent Dirichlet Allocation.

12. The computer readable storage device of claim 5, wherein the accessible content and the inaccessible content are separately stored content.