Cognitive screening of attachments

1. A computer-implemented method for cognitively processing documents to ameliorate inadvertent sharing of sensitive information during electronic communications comprising:

determining a first context, by leveraging machine learning, of an electronic communication between one or more users being prepared for transmission further comprises of comprises of, reviewing and scoring the electronic communications based on a natural language indicators of sensitive information, content of the electronic communication and recipients of the electronic communication;

determining a historical context, by leveraging machine learning, of the electronic communication;

determining, by leveraging machine learning, one or more risk scores based on the first context and the historical context;

assigning, by leveraging machine learning, the one or more risk scores to the first context and the historical context;

determining, by leveraging machine learning, whether the one or more risk scores exceed one or more predetermined thresholds based on machine learning; and

responsive to the one or more risk scores exceeding the one or more predetermined thresholds, executing, by leveraging machine learning, an action plan.

2. The computer-implemented method of claim 1 , wherein the historical context comprises of information based on previous interactions between the one or more users of the electronic communications, type of electronic communications be sent, metadata of the communications, sentiment of the metadata, natural language indicators of sensitive information, additional recipients copied on the electronic communication, evolution of the electronic communications, previous communication threads flagged as violation of internal policies, presence of locked and check-in/check out status of one or more documents relating to the electronic communications and time decay algorithm.

3. The computer-implemented method of claim 2 , wherein the time decay algorithm comprises:

determining a creation date of one or more documents;

determining last open date for the one or more documents;

determining whether new information has been added to the one or more document by comparing the current size of the one or more document against and last open date; and

determining whether the creation date, last open date and new information has been added and comparing against a pre-determined time decay table.

4. The computer-implemented method of claim 3 , wherein contextual information related to the one or more documents further comprises, presence of inline notes and comments and other documents for comparison that resides on the or more user's computer.

5. The computer-implemented method of claim 1 , wherein action plan further comprises of prompting the one or more users to remove sensitive information from the document and sending the electronic communication without the document.

6. The computer-implemented method of claim 1 , wherein one or more risk scores further comprises of file risk score (FRS), role risk score (RRS), historical risks score (HRS) and context risk score (CRS).

7. The computer-implemented method of claim 1 , wherein the one or more predetermined thresholds comprises of file risk threshold (FRT), role risk threshold (RRT), historical risk threshold (HRT) and context risk threshold (CRT).

8. A computer program product for cognitively processing documents to ameliorate inadvertent sharing of sensitive information during electronic communications, the computer program product comprising:

one or more computer readable storage media and program instructions stored on the one or more computer readable storage media, the program instructions comprising:

program instructions to determine a first context, by leveraging machine learning, of an electronic communication between one or more users being prepared for transmission further comprises of comprises of, reviewing and scoring the electronic communications based on a natural language indicators of sensitive information, content of the electronic communication and recipients of the electronic communication;

program instructions to determine a historical context, by leveraging machine learning, of the electronic communication;

program instructions to determine, by leveraging machine learning, one or more risk scores based on the first context and the historical context;

program instructions to assign, by leveraging machine learning, the one or more risk scores to the first context and the historical context;

program instructions to determine, by leveraging machine learning, whether the one or more risk scores exceed one or more predetermined thresholds based on machine learning; and

responsive to the one or more risk scores exceeding the one or more predetermined thresholds, program instructions to execute, by leveraging machine learning, an action plan.

9. The computer program product of claim 8 , wherein the historical context comprises of information based on previous interactions between the one or more users of the electronic communications, type of electronic communications be sent, metadata of the communications, sentiment of the metadata, natural language indicators of sensitive information, additional recipients copied on the electronic communication, evolution of the electronic communications, previous communication threads flagged as violation of internal policies, presence of locked and check-in/check out status of one or more documents relating to the electronic communications and time decay algorithm.

10. The computer program product of claim 9 , wherein the time decay algorithm comprises:

program instructions to determine a creation date of one or more documents;

program instructions to determine last open date for the one or more documents;

program instructions to determine whether new information has been added to the one or more document by comparing the current size of the one or more document against and last open date; and

program instructions to determine whether the creation date, last open date and new information has been added and comparing against a pre-determined time decay table.

11. The computer program product of claim 10 , wherein the determining the wherein contextual information related to the one or more documents further comprises, presence of inline notes and comments and other documents for comparison that resides on the or more user's computer.

12. The computer program product of claim 8 , wherein action plan further comprises of prompting the one or more users to remove sensitive information from the document and sending the electronic communication without the document.

13. The computer program product of claim 8 , wherein one or more risk scores further comprises of file risk score (FRS), role risk score (RRS), historical risks score (HRS) and context risk score (CRS).

14. The computer program product of claim 8 , wherein the one or more predetermined thresholds comprises of file risk threshold (FRT), role risk threshold (RRT), historical risk threshold (HRT) and context risk threshold (CRT).

15. A computer system for cognitively processing documents to ameliorate inadvertent sharing of sensitive information during electronic communications, the computer system comprising:

one or more computer processors;

one or more computer readable storage media;

program instructions stored on the one or more computer readable storage media for execution by at least one of the one or more computer processors, the program instructions comprising:

program instructions to determine a first context, by leveraging machine learning, of an electronic communication between one or more users being prepared for transmission further comprises of comprises of, reviewing and scoring the electronic communications based on a natural language indicators of sensitive information, content of the electronic communication and recipients of the electronic communication;

program instructions to determine a historical context, by leveraging machine learning, of the electronic communication;

program instructions to determine, by leveraging machine learning, one or more risk scores based on the first context and the historical context;

program instructions to assign, by leveraging machine learning, the one or more risk scores to the first context and the historical context;

program instructions to determine, by leveraging machine learning, whether the one or more risk scores exceed one or more predetermined thresholds based on machine learning; and

responsive to the one or more risk scores exceeding the one or more predetermined thresholds, program instructions to execute, by leveraging machine learning, an action plan.

16. The computer system of claim 15 , wherein the historical context comprises of information based on previous interactions between the one or more of the electronic communications, type of electronic communications be sent, metadata of the communications, sentiment of the metadata, natural language indicators of sensitive information, additional recipients copied on the electronic communication, evolution of the electronic communications, previous communication threads flagged as violation of internal policies presence of locked and check-in/check out status of one or more documents relating to the electronic communications and time decay algorithm.

17. The computer system of claim 16 , wherein the time decay algorithm comprises:

program instructions to determine a creation date of one or more documents;

program instructions to determine last open date for the one or more documents;

program instructions to determine whether new information has been added to the one or more document by comparing the current size of the one or more document against and last open date; and

program instructions to determine whether the creation date, last open date and new information has been added and comparing against a pre-determined time decay table.

18. The computer system of claim 17 , wherein the determining the wherein contextual information related to the one or more documents further comprises, presence of inline notes and comments and other documents for comparison that resides on the or more user's computer.

19. The computer system of claim 15 , wherein action plan further comprises of prompting the or more users to remove sensitive information from the document and sending the electronic communication without the document.

20. The computer system of claim 15 , wherein one or more risk scores further comprises of file risk score (FRS), role risk score (RRS), historical risks score (HRS) and context risk score (CRS).