System and method for automatic document classification and grouping based on document topic
A document handling system includes a memory and a processor, in communication with the memory, to receive first information from a first document, determine that the first document includes a first topic based on the first information, determine a first classification level of the first document, determine a first grouping of the first document, associate the first classification level and the first grouping with the first topic, receive second information from a second document, determine that the second document includes the first topic based on the second information, and modify the second document to ascribe the first classification level and the first grouping to the second document.
1. A machine comprising:
a memory; and
a processor in communication with the memory, the processor to receive first information from a first document, determine that the first document includes a first topic cluster based on the first information, determine that a database includes the first topic cluster, determine a first classification level of the first document within a hierarchical classification system, wherein access of a user to the first document is granted when the user has a first clearance level that corresponds with the first classification level and determine a first grouping of the first document within a grouping system, wherein the access to the first document is granted when the user is associated with the first grouping, wherein determining the first classification level and the first grouping are based upon the database including an association of the first classification level and the first grouping with the first topic cluster, associate the first classification level and the first grouping with the first topic cluster, receive second information from a second document, determine that the second document includes the first topic cluster based on the second information, modify the second document to ascribe the first classification level and the first grouping to the second document in response to determining that the second document includes the first topic cluster, receive third information from a third document, determine that the third document is associated with the first classification level and the first grouping, determine that the third document includes a second topic cluster based on the third information, add the third topic cluster to the database, and grant access to the second document to a member of the first group, wherein the first document is determined to include the first topic cluster based upon a language processing analysis of the first information, the second document is determined to include the first topic cluster based upon the language processing analysis of the second information, and the third document is determined to include the second topic cluster based upon the language processing analysis of the third information.
2. The machine of claim 1 , wherein the processor is further to determine that the first document comprises a web page, and, in modifying the second document, to modify a link of the web page.
3. The machine of claim 2 , wherein in modifying the link the processor is further to determine that the link includes the first topic clustor.
4. The machine of claim 3 , wherein in modifying the link the processor is further to delete the link in response to determining that the link includes the first topic clustor.
5. The document handling system of claim 3 , wherein in modifying the link the processor is further to insert code into the first document to implement a pop-up window associated with the first topic in response to determining that the link includes the first topic clustor.
6. The machine of claim 1 , the processor further to receive fourth information from a fourth document, determine that the fourth document includes the first topic based on the fourth information, and determine a second classification level of the fourth document.
7. The machine of claim 6 , the processor further to provide a first indication that the first classification level associated with the first document is different from the second classification level associated with the fourth document, and receive a second indication to modify a selected one of the first document and the fourth document in response to providing the first indication.
8. The machine of claim 7 , the processor further to modify the first document to ascribe the second classification level to the first document when the first document is selected, and to modify the fourth document a scribe the first classification level to the third document when the third document is selected.
9. The machine of claim 1 , wherein the language processing analysis includes one of a Term Frequency-Inverse Document Frequency (TF-IDF) analysis, a Latent Dirichlet Allocation (LDA) analysis and a Latent Semantic Indexing (LSI) analysis.
10. A method comprising:
receiving, at a document handling system, first information from a first document;
determining that the first document includes a first topic based on the first information;
determining that a database includes the first topic cluster;
determining a first classification level of the first document within a hierarchical classification system, wherein access of a user to the first document is granted when the user has a first clearance level that corresponds with the first classification level and determining a first grouping of the first document within a grouping system, wherein the access to the first document is granted when the user is associated with the first grouping, wherein determining the first classification level and the first grouping are based upon the database including an association of the first classification level and the first grouping with the first topic cluster;
associating the first classification level and the first grouping with the first topic;
receiving second information from a second document;
determining that the second document includes the first topic based on the second information; and
modifying the second document to ascribe the first classification level and the first grouping to the second document in response to determining that the second document includes the first topic clustor;
receiving third information from a third document;
determining that the third document is associated with the first classification level and the first grouping;
determining that the third document includes a second topic cluster based on the third information;
adding the third topic cluster to the database; and
granting access to the second document to a member of the first group, wherein the first document is determined to include the first topic cluster based upon a language processing analysis of the first information, the second document is determined to include the first topic cluster based upon the language processing analysis of the second information, and the third document is determined to include the second topic cluster based upon the language processing analysis of the third information.
11. The method of claim 10 , further comprising:
determining that the first document comprises a web page;
wherein modifying the second document further comprises modifying a link of the web page.
12. The method of claim 11 , further comprising:
determining that the link includes the first topic clustor.
13. The method of claim 12 , further comprising:
deleting the link in response to determining that the link includes the first topic clustor.
14. The method of claim 12 , further comprising:
inserting code into the first document to implement a pop-up window associated with the first topic in response to determining that the link includes the first topic clustor.
15. The method of claim 10 , further comprising:
receiving third information from a fourth document;
determining that the fourth document includes the first topic based on the fourth information; and
determining a second classification level of the fourth document.
16. The method of claim 15 , further comprising:
providing a first indication that the first classification level associated with the first document is different from the second classification level associated with the fourth document; and
receiving a second indication to modify a selected one of the first document and the fourth document in response to providing the first indication.
17. The method of claim 16 , further comprising:
modifying the first document to ascribe the second classification level to the first document when the first document is selected; and
modifying the third document to ascribe the first classification level to the third document when the third document is selected.
18. The method of claim 10 , wherein the language processing analysis includes one of a Term Frequency-Inverse Document Frequency (TF-IDF) analysis, a Latent Dirichlet Allocation (LDA) analysis and a Latent Semantic Indexing (LSI) analysis.
19. A non-transitory computer-readable medium including code for performing a process, the process comprising:
receiving first information from a first document;
determining that the first document includes a first topic cluster based on the first information;
determining that a database includes the first topic cluster;
determining a first classification level of the first document within a hierarchical classification system, wherein access to the first document is granted when a user has a first clearance level that corresponds with the first classification level and determining a first grouping of the first document within a grouping system, wherein the access to the first document is granted when the user is associated with the first grouping, wherein determining the first classification level and the first grouping are based upon the database including an association of the first classification level and the first grouping with the first topic cluster;
associating the first classification level and the first grouping with the first topic cluster;
receiving second information from a second document;
determining that the second document includes the first topic cluster based on the second information;
modifying the second document to ascribe the first classification level and the first grouping to the second document in response to determining that the second document includes the first topic cluster;
receiving third information from a third document;
determining that the third document is associated with the first classification level and the first grouping;
determining that the third document includes a second topic cluster based on the third information;
adding the third topic cluster to the database; and
granting access to the second document to a member of the first group, wherein the first document is determined to include the first topic cluster based upon a language processing analysis of the first information, the second document is determined to include the first topic cluster based upon the language processing analysis of the second information, and the third document is determined to include the second topic cluster based upon the language processing analysis of the third information.
20. The computer-readable medium of claim 19 , wherein the language processing analysis includes one of a Term Frequency-Inverse Document Frequency (TF- IDF) analysis, a Latent Dirichlet Allocation (LDA) analysis and a Latent Semantic Indexing (LSI) analysis.