Content leakage protection
Methods and systems for identifying content of interest. Accessed textual information is processed by at least one of character unification, phrase unification, and concept unification. A configured processor executes at least one predefined rule to determine whether the unified content includes certain types of information. Unified content that matches may be subject to further action such as alerts, encryption, logging, etc.
1. A method for identifying content of interest, the method comprising:
accessing, through an interface to a programmable device, a text in a first language for analysis;
applying, using a configured processor, a plurality of unification processes chained in a series configuration to create a processed text, the plurality of unification processes including at least one character unification that replaces each of at least two distinct characters in the text with a unification character token in the processed text, at least one phrase unification that replaces each of at least two distinct phrases in the text with a single unification phrase token in the processed text, and at least one concept unification, wherein the at least one concept unification replaces a first portion of the text with a concept unification token based on a second portion of the text that is adjacent to the first portion;
identifying a content of interest in the processed text by comparing, using the configured processor, the processed text including the concept unification token to at least one predefined rule that defines content of interest as a portion of the processed text wherein the concept unification token is proximal to personally identifiable information in the processed text; and
using the configured processor, preventing transmission, beyond the programmable device, of the content of interest identified in the processed text.
2. The method of claim 1 , wherein character unification replaces any of a set of similar characters with a single character.
3. The method of claim 2 , wherein the set of similar characters are logograms or phonograms.
4. The method of claim 2 , wherein the single character is selected from the set of similar characters.
5. The method of claim 1 , wherein phrase unification replaces any of a set of similar phrases with a token.
6. The method of claim 1 , wherein concept unification replaces any of a set of values with a token.
7. The method of claim 1 , wherein the at least one predefined rule includes at least one token.
8. The method of claim 1 , wherein the at least one predefined rule includes at least one text pattern matching command sequence.
9. A system for identifying content of interest, the system comprising:
an interface to a programmable device configured to access a text in a first language for analysis;
a processor in communication with the interface, the processor configured to:
apply a plurality of unification processes chained in a series configuration to create a processed text, the plurality of unification processes including at least one character unification that replaces each of at least two distinct characters in the text with a unification character token in the processed text, at least one phrase unification that replaces each of at least two distinct phrases in the text with a single unification phrase token in the processed text, and at least one concept unification, wherein the at least one concept unification replaces a first portion of the text with a concept unification token based on a second portion of the text that is adjacent to the first portion;
identify a content of interest in the processed text by comparing the processed text including the concept unification token to at least one predefined rule that defines content of interest as a portion of the processed text wherein the concept unification token is proximal to personally identifiable information in the processed text; and
prevent transmission, beyond the programmable device, of the content of interest identified in the processed text; and
a data store in communication with the processor, the data store containing the at least one predefined rule.
10. The system of claim 9 , wherein character unification replaces any of a set of similar characters with a single character.
11. The system of claim 10 , wherein the set of similar characters are logograms or phonograms.
12. The system of claim 10 , wherein the single character is selected from the set of similar characters.
13. The system of claim 9 , wherein phrase unification replaces any of a set of similar phrases with a token.
14. The system of claim 9 , wherein concept unification replaces any of a set of values with a token.
15. The system of claim 9 , wherein the at least one predefined rule includes at least one token.
16. The system of claim 9 , wherein the at least one predefined rule includes at least one regular expression.
17. The system of claim 9 , wherein at least one of the unification character token and the unification phrase token comprises a second language that is distinct from the first language.
18. A non-transitory computer readable medium containing computer-executable instructions for performing a method for identifying content of interest, the non-transitory computer readable medium comprising:
computer-executable instructions for accessing, through an interface to a programmable device, a text in a first language for analysis;
computer-executable instructions for applying, using a configured processor, a plurality of unification processes chained in a series configuration to create a processed text, the plurality of unification processes including at least one character unification that replaces each of at least two distinct characters in the text with a unification character token in the processed text, at least one phrase unification that replaces each of at least two distinct phrases in the text with a single unification phrase token in the processed text, and at least one concept unification, wherein the at least one concept unification replaces a first portion of the text with a concept unification token based on a second portion of the text that is adjacent to the first portion;
computer-executable instructions for identifying a content of interest in the processed text by comparing, using the configured processor, the processed text including the concept unification token to at least one predefined rule that defines content of interest as a portion of the processed text wherein concept unification token is proximal to personally identifiable information in the processed text, the at least one predefined rule further including a regular expression; and
computer-executable instructions for preventing transmission, beyond the programmable device, of the content of interest identified in the processed text.
19. The non-transitory computer readable medium of claim 18 , wherein at least one of the unification character token and the unification phrase token comprises a second language that is distinct from the first language.