Language independent probabilistic content matching

• US 9,633,001 B2
• Filed: 06/23/2015
• Issued: 04/25/2017
• Est. Priority Date: 02/07/2012
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Gandhi, Mauktik, Lamanna, Charles, Sankaranarayanan, Vidyaraman, Pontes Filho, Raimundo
• Primary Examiner(s): JACKSON, JAKIEDA R
• Application Number:
  US14/747,428
• Publication Number:
  US 20160012037A1
• Time in Patent Office: 672 Days
• Field of Search: 704244, 704 9
• CPC Class Codes: G06F 21/6263 during internet communication, e.g. revealing personal data from cookies
• CPC Class Codes: G06F 40/166 Editing, e.g. inserting or deleting
• CPC Class Codes: G06F 40/20 Natural language analysis semantic analysis of natural language G06F40/30
• CPC Class Codes: G06F 40/279 Recognition of textual entities
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: H04L 63/10 for controlling access to devices or network resources

Abstract

Content is received and compared against rules for identifying a type of content. Each rule has both segmented and unsegmented patterns. The content is matched against the patterns and assigned a confidence score that is higher if the content matches a segmented pattern and lower if the content matches an unsegmented pattern.

Claims

1. A computing system comprising:

at least one processor; and

memory storing instructions executable by the at least one processor, wherein the instructions configure the computing system;

access a rule that defines patterns that are used to identify content as sensitive content, the rule defining

a segmented pattern to be matched to textual content written in a segmented language, and corroborating data associated with the segmented pattern, and

an un-segmented pattern to be matched to textual content written in an un-segmented language, and corroborating data associated with the un-segmented pattern;

identify an electronic source document having electronic document content;

determine whether the electronic document content is sensitive content by matching the electronic document content against the patterns in the rule and generating a confidence score corresponding to the determination as to whether the electronic document content is sensitive content, wherein generation of the confidence score is based on whether the electronic document content matched the segmented pattern or the un-segmented pattern, and based on the corroborating data associated with the matched pattern, the generation of the confidence score being regardless of a language of the electronic document content;

identify a data dissemination policy based on the determination as to whether the electronic document content is sensitive content and the corresponding confidence score; and

automatically process the electronic document by identifying an action defined by the data dissemination policy and automatically performing the identified action to control electronic dissemination of the electronic document content over a computer network by at least one of;

automatically blocking the document content from being sent to a potential recipient;

automatically displaying a message indicating that the document content contains sensitive material and that the document content will be blocked from being sent to a potential recipient;

or

automatically displaying a message indicating that the document content contains sensitive material and instructing the user how to proceed based on the data dissemination policy.

2. The computing system of claim 1, wherein the instructions configure the computing system to assign a higher confidence level to the determination if the electronic document content matched a segmented pattern than if the electronic document content matched an un-segmented pattern.

3. The computing system of claim 2, wherein the rule comprises a main matching portion and a corroborating matching portion, and wherein the instructions configure the computing system to match the electronic document content first against the main matching portion and then, if the electronic document content matched the main matching portion, against the corroborating matching portion.

4. The computing system of claim 3, wherein the instructions configure the computing system to assign an increased confidence level it the electronic document content matches more than one of the plurality matching patterns in the corroborating matching portion.

5. The computing of claim 1, wherein the instructions configure the computing system to match the document content against the patterns in the rule that identify at least one of:

personal information, credit card information, or social security number information.

6. The computing system of claim 1, wherein the instructions configure the computing system to:

access a plurality of rules for a plurity of different types of sensitive content, wherein each rule corresponds to a particular of the different type of sensitive content and defines;

a segmented pattern to be matched to textual content written in a segmented language, and corroborating data associated with the segmented pattern, and

an un-segmented pattern to be matched to textual content written in an un-segmented language, and corroborating data associated with the un-segmented pattern.

7. The computing system of claim 6, wherein the instructions configure computing system to:

determine whether the electronic document content is sensitive content by searching and applying each of the plurality of rules to the electronic document content regardless of the language of the electronic document content.

8. The computing system of claim 1, wherein the un-segment pattern defined in each rule identifies a set of associated keywords in a plurality of different un-segmented languages.

9. A computing system comprising:

a processor; and

memory storing instructions which, when executed by the processor, configure the computing system to;

access a rule for identifying a particular content type, the rule comprising;

a segmented pattern that, when matched to textual content written in a segmented language, defines the textual content as the particular content type; and

an unsegmented pattern that, when matched to textual content written in an unsegmented language, defines the textual content as the particular content type;

identify an electronic message having message content;

generate a confidence score indicative of whether the message content is of the particular content type, the confidence score being generated by applying the rule to the message content and determining whether the message content matched the segmented pattern or the un-segmented pattern; and

based on the confidence score, process the electronic message by at least one of;

blocking the message content from being sent to a potential recipient;

rendering a user notification indicating that the message content contains sensitive material and will be blocked from being sent to a potential recipient; and

rendering a user notification indicating that the message content contains sensitive material and instructing the user how to proceed to comply with a data dissemination policy.

10. The computing system of claim 9, wherein the electronic message comprises an email message and the message content comprises at least one of:

an email body or an email attachment.