Natural language processing system to analyze mobile application feedback

• US 10,162,812 B2
• Filed: 04/04/2017
• Issued: 12/25/2018
• Est. Priority Date: 04/04/2017
• Status: Active Grant

• Current Assignee: Bank of America Corp.
• Original Assignee: Bank of America Corp.
• Inventors: Pradhan, Satyabrata, Bhashetty, Sunil, Gottumukkala, Raja Venkatesh
• Primary Examiner(s): Pullias, Jesse S
• Application Number:
  US15/478,769
• Publication Number:
  US 20180285345A1
• Time in Patent Office: 630 Days
• Field of Search: 704 1, 704 9, 704 10, 704257, 704270-275
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

Aspects of the disclosure relate to using a natural language processing system to analyze mobile application feedback. A computing platform having at least one processor, a memory, and a communication interface may receive mobile application feedback information comprising text feedback associated with feedback of a mobile application. The computing platform may identify one or more nouns associated with the text feedback. The computing platform may identify one or more text feedback topics. The computing platform may generate one or more commands directing a sentiment analysis server to determine one or more sentiments for the one or more text feedback topics. The computing platform may transmit the one or more commands directing the sentiment analysis server to determine the one or more sentiments. The computing platform may receive the one or more sentiments. The computing platform may transmit the feedback topics and the one or more sentiments.

Claims

1. A computing platform comprising:

at least one processor;

a communication interface communicatively coupled to the at least one processor; and

memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to;

receive, by the at least one processor, via the communication interface, and from a first user device, mobile application feedback information comprising text feedback associated with feedback of a mobile application;

identify, based on the text feedback, one or more nouns associated with the text feedback;

identify, based on a comparison between the one or more nouns with a plurality of mobile application topics associated with the mobile application, one or more text feedback topics;

generate, based on the one or more text feedback topics, one or more commands directing a sentiment analysis server to determine one or more sentiments for the one or more text feedback topics, wherein the generating the one or more commands directing the sentiment analysis server to determine the one or more text feedback topics comprises;

determining, based on performing sentiment analysis on a part of the text feedback associated with the one or more text feedback topics, the one or more sentiments for the one or more text feedback topics, wherein determining the one or more sentiments for the one or more text feedback topics comprises;

receiving a sentiment analysis model comprising past recorded user feedback data, and

determining, based on a comparison between the sentiment analysis model and the part of the text feedback associated with the one or more text feedback topics, the one or more sentiments and one or more score probabilities associated with the one or more sentiments, and

transmitting the one or more sentiments for the one or more text feedback topics;

transmit, via the communication interface and to the sentiment analysis server, the one or more commands directing the sentiment analysis server to determine the one or more sentiments;

receive, via the communication interface and from the sentiment analysis server, the one or more sentiments; and

transmit, via the communication interface and to a summarization server, the one or more text feedback topics and the one or more sentiments.

2. The computing platform of claim 1, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:

modify, using a lemmatization analysis technique, the text feedback to generate cleansed text feedback; and

wherein the one or more nouns associated with the text feedback is identified based on the cleansed text feedback.

3. The computing platform of claim 1, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:

receive, via the communication interface and from a mobile application dictionary server, the plurality of mobile application topics associated with the mobile application; and

wherein the identifying the one or more text feedback topics is based on the one or more nouns matching one or more mobile application topics from the plurality of mobile application topics.

4. The computing platform of claim 3, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:

responsive to the one or more nouns not matching the plurality of mobile application topics, determine a frequency of user feedback associated with the one or more nouns and a time period associated with the one or more nouns; and

identify, based on the frequency of user feedback and the time period, new mobile application topics associated with the mobile application.

5. The computing platform of claim 4, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:

modify the plurality of mobile application topics associated with the mobile application to include the new mobile application topics; and

transmit, via the communication interface and to the mobile application dictionary server, the plurality of modified mobile application topics.

6. The computing platform of claim 1, wherein the determining the one or more sentiments comprises:

extracting, based on using a natural language processing based dependency parser, the part of the text feedback associated with the one or more text feedback topics; and

determining the one or more sentiments from the part of the text feedback associated with the one or more text feedback topics.

7. The computing platform of claim 1, wherein the determining the one or more score probabilities is based on similarities between the part of the text feedback associated with the one or more text feedback topics and phrases within the sentiment analysis model.

8. The computing platform of claim 1, wherein the mobile application feedback information further comprises a voice recording associated with feedback of the mobile application;

and

wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to generate the text feedback from the voice recording.

9. The computing platform of claim 1, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:

receive, by the at least one processor, via the communication interface, and from a plurality of user devices, second mobile application feedback information comprising a plurality of text feedback associated with feedback of the mobile application;

identify, based on the plurality of text feedback, a plurality of nouns associated with the plurality of text feedback;

identify, based on a comparison between the plurality of nouns with the plurality of mobile application topics associated with the mobile application, a plurality of text feedback topics associated with the plurality of text feedback;

generate, based on the plurality of text feedback topics associated with the plurality of text feedback, one or more commands directing the sentiment analysis server to determine a plurality of sentiments for the plurality of text feedback topics associated with the plurality of text feedback;

transmit, via the communication interface and to the sentiment analysis server, the one or more commands directing the sentiment analysis server to determine the plurality of sentiments, wherein each of the plurality of sentiments corresponds to a text feedback topic from the plurality of text feedback topics associated with the plurality of text feedback;

receive, via the communication interface and from the sentiment analysis server, the plurality of sentiments; and

transmit, via the communication interface and to the summarization server, the plurality of text feedback topics associated with the plurality of text feedback and the plurality of sentiments.

10. The computing platform of claim 9, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:

aggregate the plurality of text feedback topics associated with the plurality of text feedback and the plurality of sentiments;

determine one or more extracted comments from the plurality of text feedback associated with popular user feedback comments corresponding to the mobile application;

generate, based on the aggregated plurality of text feedback topics associated with the plurality of text feedback, the aggregated plurality of sentiments, and the one or more extracted comments, a report associated with the second mobile application feedback information; and

transmit, via the communication interface and to the summarization server, the report.

11. A method comprising:

at a computing platform comprising at least one processor, memory, and a communication interface;

receiving, by the at least one processor, via the communication interface, and from a first user device, mobile application feedback information comprising text feedback associated with feedback of a mobile application;

identifying, by the at least one processor, based on the text feedback, one or more nouns associated with the text feedback;

identifying, by the at least one processor and based on a comparison between the one or more nouns with a plurality of mobile application topics associated with the mobile application, one or more text feedback topics;

generating, by the at least one processor and based on the one or more text feedback topics, one or more commands directing a sentiment analysis server to determine one or more sentiments for the one or more text feedback topics, wherein the generating the one or more commands directing the sentiment analysis server to determine the text feedback topics comprises;

determining, by the at least one processor and based on performing sentiment analysis on a part of the text feedback associated with the one or more text feedback topics, the one or more sentiments for the one or more text feedback topics, wherein determining the one or more sentiments comprises;

receiving, by the at least one processor, a sentiment analysis model comprising past recorded user feedback data, and

determining, by the at least one processor and based on a comparison between the sentiment analysis model and the part of the text feedback associated with the one or more text feedback topics, the one or more sentiments and one or more score probabilities associated with the one or more sentiments, and

transmitting the one or more sentiments for the one or more text feedback topics;

transmitting, by the at least one processor, via the communication interface, and to the sentiment analysis server, the one or more commands directing the sentiment analysis server to determine the one or more sentiments;

receiving, by the at least one processor, via the communication interface, and from the sentiment analysis server, the one or more sentiments; and

transmitting, by the at least one processor, via the communication interface, and to a summarization server, the one or more text feedback topics and the one or more sentiments.

12. The method of claim 11, further comprising:

modifying, by the at least one processor and using a lemmatization analysis technique, the text feedback to generate cleansed text feedback; and

wherein the one or more nouns associated with the text feedback is identified based on the cleansed text feedback.

13. The method of claim 11, further comprising:

receiving, by the at least one processor, via the communication interface, and from a mobile application dictionary server, the plurality of mobile application topics associated with the mobile application; and

wherein the identifying the one or more text feedback topics is based on the one or more nouns matching one or more mobile application topics from the plurality of mobile application topics.

14. The method of claim 11, wherein the determining the one or more sentiments comprises:

extracting, by the at least one processor and based on using a natural language processing based dependency parser, the part of the text feedback associated with the one or more text feedback topics; and

determining, by the at least one processor, the one or more sentiments from the part of the text feedback associated with the one or more text feedback topics.

15. The method of claim 11, wherein the determining the one or more score probabilities is based on similarities between the part of the text feedback associated with the one or more text feedback topics and phrases within the sentiment analysis model.

16. One or more non-transitory computer-readable media storing instructions that, when executed by a computing platform comprising at least one processor, memory, and a communication interface, cause the computing platform to:

receive, by the at least one processor, via the communication interface, and from a first user device, mobile application feedback information comprising text feedback associated with feedback of a mobile application;

identify, based on the text feedback, one or more nouns associated with the text feedback;

identify, based on a comparison between the one or more nouns with a plurality of mobile application topics associated with the mobile application, one or more text feedback topics;

generate, based on the one or more text feedback topics, one or more commands directing a sentiment analysis server to determine one or more sentiments for the one or more text feedback topics, wherein the generating the one or more commands directing the sentiment analysis server to determine the one or more text feedback topics comprises;

determining, based on performing sentiment analysis on a part of the text feedback associated with the one or more text feedback topics, the one or more sentiments for the one or more text feedback topics, wherein determining the one or more sentiments for the one or more text feedback topics comprises;

receiving a sentiment analysis model comprising past recorded user feedback data, and

determining, based on a comparison between the sentiment analysis model and the part of the text feedback associated with the one or more text feedback topics, the one or more sentiments and one or more score probabilities associated with the one or more sentiments, and

transmitting the one or more sentiments for the one or more text feedback topics;

transmit, via the communication interface and to the sentiment analysis server, the one or more commands directing the sentiment analysis server to determine the one or more sentiments;

receive, via the communication interface and from the sentiment analysis server, the one or more sentiments; and

transmit, via the communication interface and to a summarization server, the one or more text feedback topics and the one or more sentiments.

17. The one or more non-transitory computer-readable media of claim 16, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:

modify, using a lemmatization analysis technique, the text feedback to generate cleansed text feedback; and

wherein the one or more nouns associated with the text feedback is identified based on the cleansed text feedback.

18. The one or more non-transitory computer-readable media of claim 16, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:

receive, via the communication interface and from a mobile application dictionary server, the plurality of mobile application topics associated with the mobile application; and

wherein the identifying the one or more text feedback topics is based on the one or more nouns matching one or more mobile application topics from the plurality of mobile application topics.

19. The one or more non-transitory computer-readable media of claim 18, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:

responsive to the one or more nouns not matching the plurality of mobile application topics, determine a frequency of user feedback associated with the one or more nouns and a time period associated with the one or more nouns; and

identify, based on the frequency of user feedback and the time period, new mobile application topics associated with the mobile application.

20. The one or more non-transitory computer-readable media of claim 19, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:

modify the plurality of mobile application topics associated with the mobile application to include the new mobile application topics; and

transmit, via the communication interface and to the mobile application dictionary server, the plurality of modified mobile application topics.