METHOD AND A SYSTEM TO SUMMARIZE A CONVERSATION

• US 20180189266A1
• Filed: 02/15/2017
• Published: 07/05/2018
• Est. Priority Date: 01/03/2017
• Status: Active Application

• Current Assignee: Wipro Limited
• Original Assignee: Wipro Limited
• Application Number:
  US15/433,687
• Publication Number:
  US 20180189266A1
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/345 Summarisation for human users
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/35 Discourse or dialogue representation

Abstract

A method and a system are described for summarizing a conversation. The method includes splitting a plurality of sentences in the conversation into one or more phrases based on pre-defined grammar rules. The method includes extracting one or more relevant phrases associated with each of the one or more phrases based on domain dictionaries and pre-defined knowledge base. The method includes determining a correlation between each of the one or more phrases and the one or more relevant phrases based on a relationship score. The method includes identifying at least one of a core topic of discussion in the conversation, one or more issues, one or more resolved issues, and one or more unresolved issues based on the correlation.

The method includes creating a summary of the conversation based on the core topic of discussion, one or more issues, one or more resolved issues, and one or more unresolved issues.

Claims

1. A method of summarizing a conversation, the method comprising:

splitting, by a summary creation server, each of a plurality of sentences in the conversation into one or more phrases based on one or more pre-defined grammar rules, wherein the one or more phrases are utilized to update a pre-defined knowledge base;

extracting, by the summary creation server, one or more relevant phrases associated with each of the one or more phrases based on domain dictionaries and the pre-defined knowledge base;

determining, by the summary creation server, a correlation between each of the one or more phrases and the one or more relevant phrases based on a relationship score;

identifying, by the summary creation server, at least one of a core topic of discussion in the conversation, one or more issues associated with the core topic of discussion, one or more resolved issues from the one or more issues, and one or more unresolved issues from the one or more issues based on the determined correlation; and

creating, by the summary creation server, a summary of the conversation based on the identified core topic of discussion in the conversation, the one or more issues, the one or more resolved issues, and the one or more unresolved issues.

2. The method of claim 1, further comprising removing dis-ambiguity within each of the plurality of sentences based on the one or more pre-defined grammar rules.

3. The method of claim 1, further comprising identifying one or more anaphors within each of the plurality of sentences, wherein each of the one or more anaphors are replaced with corresponding one or more nouns.

4. The method of claim 1, further comprising removing one or more non-American Standard Code for Information Interchange (ASCII) characters from each of the plurality of sentences.

5. The method of claim 1, wherein the pre-defined knowledge base comprises the one or more phrases, that indicate at least one of:

an issue, a resolution of the issue, or one or more action items associated with the issue.

6. The method of claim 1, wherein the one or more relevant phrases are associated with a domain of the one or more phrases.

7. The method of claim 1, wherein identifying the core topic of discussion in the conversation further comprises:

determining a type of each of the plurality of sentences based on one or more pre-defined sentence parsing techniques, wherein the type corresponds to one of a question and an answer,

identifying a question-answer pair based on the type of each of the plurality of sentences and one or more logical parsing rules.

8. A summary creation server comprising:

a processor; and

a memory communicatively coupled to the processor, wherein the memory stores processor instructions, which, on execution, causes the processor to;

split each of a plurality of sentences in the conversation into one or more phrases based on one or more pre-defined grammar rules, wherein the one or more phrases are utilized to update a pre-defined knowledge base;

extract one or more relevant phrases associated with each of the one or more phrases based on domain dictionaries and the pre-defined knowledge base;

determine a correlation between each of the one or more phrases and the one or more relevant phrases based on a relationship score;

identify at least one of a core topic of discussion in the conversation, one or more issues associated with the core topic of discussion, one or more resolved issues from the one or more issues, and one or more unresolved issues from the one or more issues based on the determined correlation; and

create a summary of the conversation based on the identified core topic of discussion in the conversation, the one or more issues, the one or more resolved issues, and the one or more unresolved issues.

9. The summary creation server of claim 8, wherein the processor is further configured to remove dis-ambiguity within each of the plurality of sentences based on the one or more pre-defined grammar rules.

10. The summary creation server of claim 8, wherein the processor is further configured to identify one or more anaphors within each of the plurality of sentences, wherein each of the one or more anaphors are replaced with corresponding one or more nouns.

11. The summary creation server of claim 8, wherein the processor is further configured to remove one or more non-American Standard Code for Information Interchange (ASCII) characters from each of the plurality of sentences.

12. The summary creation server of claim 8, wherein the pre-defined knowledge base comprises the one or more phrases, that indicate at least one of an issue, a resolution of the issue, or one or more action items associated with the issue.

13. The summary creation server of claim 8, wherein the one or more relevant phrases are associated with a domain of the one or more phrases.

14. The summary creation server of claim 8, wherein identifying the core topic of discussion in the conversation further comprises:

determining a type of each of the plurality of sentences based on one or more pre-defined sentence parsing techniques, wherein the type corresponds to one of a question and an answer,

identifying a question-answer pair based on the type of each of the plurality of sentences and one or more logical parsing rules.

15. A non-transitory computer-readable storage medium having stored thereon, a set of computer-executable instructions for causing a computer comprising one or more processors to perform steps comprising:

splitting each of a plurality of sentences in a conversation into one or more phrases based on one or more pre-defined grammar rules, wherein the one or more key phrases are utilized to build a pre-defined knowledge base;

extracting one or more relevant phrases associated with each of the one or more phrases based on domain dictionaries and the pre-defined knowledge base;

determining a correlation between each of the one or more phrases and the one or more relevant phrases based on a relationship score;

identifying at least one of a core topic of discussion in the conversation, one or more issues associated with the core topic of discussion, one or more resolved issues from the one or more issues, and one or more unresolved issues from the one or more issues based on the determined correlation; and

creating a summary of the conversation based on the identified core topic of discussion in the conversation, the one or more issues, the one or more resolved issues, and the one or more unresolved issues.