System for hiding sensitive messages within non-sensitive meaningful text

1. A computer-implemented method for hiding confidential text within non-confidential text, the method comprising:

receiving, by one or more computer processors, a text that includes (i) a first text and (ii) a second text, wherein at least a portion of the received first text is confidential and at least a portion of the received second text is non-confidential;

identifying, by one or more computer processors, an intended recipient of the received text;

determining, by one or more computer processors, one or more entities contained within (i) the received first text and (ii) the received second text;

analyzing, by one or more computer processors, the one or more entities, wherein the one or more entities comprise one or more topics;

identifying, by one or more computer processors, a first conversion model, which corresponds to text between the intended recipient and a creator of the text, wherein the conversion model is based on a non-confidential topic contained within the second text as well as the intended recipient; and

converting, by one or more computer processors, the received first text into a third text that has a similar topic to the non-confidential second text topic, wherein converting replaces a confidential first text with coherent non-confidential text associated with the non-confidential topic.

2. The computer-implemented method of claim 1 , the method further comprising:

receiving, by the one or more computer processors, the text that includes, one or a combination of, (i) the confidential first text and (ii) the non-confidential second text;

analyzing, by the one or more computer processors, the received text; and

identifying, by the one or more computer processors, (i) the identified recipient of the text and (ii) the creator of the text.

3. The computer-implemented method of claim 1 , the method further comprising:

identifying, by the one or more computer processors, the text document;

identifying, by the one or more computer processors, (i) one or more confidential first text and (ii) one or more non-confidential second text contained within the text document;

separating, by the one or more computer processors, the identified text document into (i) one or more portions of confidential first text and (ii) one or more portions of non-confidential second text;

classifying, by the one or more computer processors, (i) the one or more portions of confidential first text into one or more topics and (ii) the one or more portions of non-confidential second text into one or more topics; and

generating, by the one or more computer processors, one or more text conversion models based upon, one or a combination of, (i) the separated one or more portions of confidential first text, (ii) the separated one or more portions of non-confidential text, and (iii) the classified one or more topics, wherein the one or more text conversion models use a statistical language model to create one or more versions of converted text.

4. The computer-implemented method of claim 3 , wherein one or more topics are generic classifications for types of related text.

5. The computer-implemented method of claim 1 , wherein converting, by the one or more computer processors, the received confidential first text into the third text that includes a non-confidential topic that is associated with the non-confidential second text based upon the identified first conversion model includes, one or a combination of:

identifying, by the one or more computer processors, a non-confidential topic for the third text associated with the non-confidential second text; and

converting, by the one or more computer processors, the received first text into the third text that includes a non-confidential topic that is associated with the non-confidential second text based upon the identified first conversion model and the identified non-confidential topic associated with the non-confidential second text.

6. The computer-implemented method of claim 1 , wherein converting the received confidential first text into a third text that includes a non-confidential topic that is associated with the non-confidential second text based upon the identified first conversion model further includes, one or a combination of:

identifying, by the one or more computer processors, one or more entities, relations, co-references, and events included in the received confidential first text; and

converting, by the one or more computer processors, the one or more entities, relations, co-references, and events into the third text that includes a non-confidential topic that is associated with the non-confidential second text based upon the identified first conversion model.

7. The computer-implemented method of claim 2 , wherein converting (i) the third text and (ii) the second text into an original coherent message based upon a second conversion model that includes, one or a combination of:

identifying, by the one or more computer processors, the creator corresponding to the third text;

identifying, by the one or more computer processors, a second conversion model that corresponds to text received from the creator;

determining, by the one or more computer processors, that the non-confidential portion of the third text includes converted confidential first text; and

generating, by the one or more computer processors, an original message that includes, one or a combination of, (i) identified confidential first text that is converted into non- confidential third text associated with the non-confidential second text topic, (ii) identified non-confidential second text, and (iii) the identified second conversion model, which correspond to text received from the creator.

8. A computer program product for hiding confidential text within non-confidential text, the computer program product comprising:

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

program instructions to receive the text that includes (i) a first text and (ii) a second text, wherein at least a portion of the received first text is confidential and at least a portion of the received second text is non-confidential;

program instructions to identify an intended recipient of the received text;

program instructions to determine one or more entities contained within (i) the received first text and (ii) the received second text;

program instructions to analyze the one or more entities, wherein the one or more entities comprise one or more topics;

program instructions to identify a first conversion model, which corresponds to text between the intended recipient and a creator of the text, wherein the conversion model is based on a non-confidential topic contained within the second text as well as the intended recipient; and

program instructions to convert the received first text into a third text that has a similar topic to the non-confidential second text topic, wherein converting replaces a confidential first text with coherent non-confidential text associated with the non-confidential topic.

9. The computer program product of claim 8 , the program instructions further comprising:

program instructions to receive the text that includes, one or a combination of, (i) the confidential first text and (ii) the non-confidential second text;

program instructions to analyze the received text; and

program instructions to identify (i) the identified recipient of the text and (ii) the creator of the text.

10. The computer program product of claim 8 , the program instructions further comprising:

program instruction to identify a text document;

program instructions to identify (i) one or more confidential first text and (ii) one or more non-confidential second text contained within the text document;

program instructions to separate the identified text document into (i) one or more portions of confidential first text and (ii) one or more portions of non-confidential second text;

program instructions to classify (i) the one or more portions of confidential first text into one or more topics and (ii) the one or more portions of non-confidential second text into one or more topics; and

program instructions to generate one or more text conversion models based upon, one or a combination of, (i) the separated one or more portions of confidential first text, (ii) the separated one or more portions of non-confidential text, and (iii) the classified one or more topics, wherein the one or more text conversion models use a statistical language model to create one or more versions of converted text.

11. The computer program product of claim 10 , wherein one or more topics are generic classifications for types of related text.

12. The computer program product of claim 8 , wherein converting, by the one or more computer processors, the received confidential first text into the third text that includes a non-confidential topic that is associated with the non-confidential second text based upon the identified first conversion model includes, one or a combination of:

program instructions to identify a non-confidential topic for the third text associated with the non-confidential second text; and

program instructions to convert the received first text into the third text that includes a non-confidential topic that is associated with the non-confidential second text based upon the identified first conversion model and the identified non-confidential topic associated with the non-confidential second text.

13. The computer program product of claim 8 , wherein converting the received confidential first text into a third text that includes a non-confidential topic that is associated with the non-confidential second text based upon the identified first conversion model further includes, one or a combination of:

program instructions to identify one or more entities, relations, co-references, and events included in the received confidential first text; and

program instructions to convert the one or more entities, relations, co-references, and events into the third text that includes a non-confidential topic that is associated with the non- confidential second text based upon the identified first conversion model.

14. The computer program product of claim 9 , wherein converting (i) the third text and (ii) the second text into an original coherent message based upon a second conversion model that includes, one or a combination of:

program instructions to identify the creator corresponding to the third text;

program instructions to identify a second conversion model that corresponds to text received from the creator;

program instructions to determine that the non-confidential portion of the third text includes converted confidential first text; and

program instructions to generate an original message that includes, one or a combination of, (i) identified confidential first text that is converted into non-confidential third text associated with the non-confidential second text topic, (ii) identified non-confidential second text, and (iii) the identified second conversion model, which correspond to text received from the creator.

15. A computer system for hiding confidential text within non-confidential text, the computer system comprising:

one or more computer processors;

one or more computer readable storage medium; and

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

program instructions to receive a text that includes (i) a first text and (ii) a second text, wherein at least a portion of the received first text is confidential and at least a portion of the received second text is non-confidential;

program instructions to identify an intended recipient of the received text;

program instructions to determine one or more entities contained within (i) the received first text and (ii) the received second text;

program instructions to analyze the one or more entities, wherein the one or more entities comprise one or more topics;

program instructions to identify a first conversion model, which corresponds to text between the intended recipient and a creator of the text, wherein the conversion model is based on a non-confidential topic contained within the second text as well as the intended recipient; and

program instructions to convert the received first text into a third text that has a similar topic to the non-confidential second text topic, wherein converting replaces a confidential first text with coherent non-confidential text associated with the non-confidential topic.

16. The computer system of claim 15 , the program instructions further comprising:

program instructions to receive the text that includes, one or a combination of, (i) the confidential first text and (ii) the non-confidential second text;

program instructions to analyze the received text; and

program instructions to identify (i) the identified recipient of the text and (ii) the creator of the text.

17. The computer system of claim 15 , the program instructions further comprising:

program instruction to identify the text document;

program instructions to identify (i) one or more confidential first text and (ii) one or more non-confidential second text contained within the text document;

program instructions to separate the identified text document into (i) one or more portions of confidential first text and (ii) one or more portions of non-confidential second text;

program instructions to classify (i) the one or more portions of confidential first text into one or more topics and (ii) the one or more portions of non-confidential second text into one or more topics; and

program instructions to generate one or more text conversion models based upon, one or a combination of, (i) the separated one or more portions of confidential first text, (ii) the separated one or more portions of non-confidential text, and (iii) the classified one or more topics, wherein the one or more text conversion models use a statistical language model to create one or more versions of converted text.

18. The computer system of claim 15 , wherein converting, by the one or more computer processors, the received confidential first text into the third text that includes a non-confidential topic that is associated with the non-confidential second text based upon the identified first conversion model includes, one or a combination of:

program instructions to identify a non-confidential topic for the third text associated with the non-confidential second text; and

program instructions to convert the received first text into the third text that includes a non-confidential topic that is associated with the non-confidential second text based upon the identified first conversion model and the identified non-confidential topic associated with the non-confidential second text.

19. The computer system of claim 15 , wherein converting the received confidential first text into a third text that includes a non-confidential topic that is associated with the non-confidential second text based upon the identified first conversion model further includes, one or a combination of:

program instructions to identify one or more entities, relations, co-references, and events included in the received confidential first text; and

program instructions to convert the one or more entities, relations, co-references, and events into the third text that includes a non-confidential topic that is associated with the non-confidential second text based upon the identified first conversion model.

20. The computer system of claim 16 , wherein converting (i) the third text and (ii) the second text into an original coherent message based upon a second conversion model that includes, one or a combination of:

program instructions to identify the creator corresponding to the third text;

program instructions to identify a second conversion model that corresponds to text received from the creator;

program instructions to determine that the non-confidential portion of the third text includes converted confidential first text; and

program instructions to generate an original message that includes, one or a combination of, (i) identified confidential first text that is converted into non-confidential third text associated with the non-confidential second text topic, (ii) identified non-confidential second text, and (iii) the identified second conversion model, which correspond to text received from the creator.