Time-based sentiment analysis for product and service features

• US 9,177,554 B2
• Filed: 02/04/2013
• Issued: 11/03/2015
• Est. Priority Date: 02/04/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bhatt, Dhruv A., McNeil, Kristin E., Patel, Nitaben A.
• Primary Examiner(s): Le, Vu
• Assistant Examiner(s): WOLDEMARIAM, AKLILU K
• Application Number:
  US13/758,560
• Publication Number:
  US 20140219571A1
• Time in Patent Office: 1,002 Days
• Field of Search: 382/209, 382/224, 382/229, 382/305, 707/608, 707/706, 707/730, 707/748, 704/10, 704/9, 704/1
• US Class Current: 1/1
• CPC Class Codes: G06F 16/36 Creation of semantic tools, e.g. ontology or thesauri
• CPC Class Codes: G06F 16/38 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 40/00 Handling natural language data speech analysis or synthesis, speech recognition G10L
• CPC Class Codes: G06F 40/237 Lexical tools
• CPC Class Codes: G06Q 30/00 Commerce
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06V 10/768 using context analysis, e.g. recognition aided by known co-occurring patterns
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

Provided are a method, computer program product and system for reporting time-based sentiment for a product. Text analysis is performed on at least one communication. At least one feature for the product is determined based on the text analysis. A sentiment value is generated for the at least one feature for the product. A date associated with the sentiment value is determined, and the sentiment value is reported for at least one feature over time.

Claims

1. A computer implemented method for reporting sentiment of a product, comprising:

performing a text analysis on communications, by at least one computing processor;

determining at least one feature for the product based on the text analysis, by the at least one computing processor;

generating sentiment values using the communications for the at least one feature for the product based on a sentiment dictionary and sentiment rules that determine a sentiment strength, by the at least one computing processor;

determining a date associated with each of the sentiment values by extracting the date from the communications, by the at least one computing processor;

for each date associated with each of the sentiment values, recording a feature annotation, a sentiment annotation, the sentiment value, metadata, and the date, wherein the feature annotation is generated using a feature dictionary and feature rules, and wherein the sentiment annotation is generated using the sentiment dictionary and the sentiment rules, by the at least one computing processor; and

reporting how the sentiment values changed over time based on each date, by the at least one computing processor.

2. The method of claim 1, wherein generating a sentiment value from the sentiment values includes determining at least one sentiment for the at least one feature based on the text analysis.

3. The method of claim 1, wherein the product is a service.

4. The method of claim 1, further comprising:

identifying an audio communication; and

converting the audio communication to text.

5. The method of claim 1, wherein each date is selected from:

a purchase date; and

a communication date.

6. The method of claim 2, wherein generating the sentiment value includes combining a plurality of sentiments for the at least one feature.

7. A system for reporting sentiment of a product, comprising:

a processor; and

a storage device connected to the processor, wherein the storage device has stored thereon a program, wherein the processor is configured to execute instructions of the program to perform operations, and wherein the operations comprise;

performing a text analysis on communications;

determining at least one feature for the product based on the text analysis;

generating the sentiment values using the communications for the at least one feature for the product based on a sentiment dictionary and sentiment rules that determine a sentiment strength;

determining a date associated with each of the sentiment values by extracting the date from the communications;

for each date associated with each of the sentiment values, recording a feature annotation, a sentiment annotation, the sentiment value, metadata, and the date, wherein the feature annotation is generated using a feature dictionary and feature rules, and wherein the sentiment annotation is generated using the sentiment dictionary and the sentiment rules; and

reporting how the sentiment values changed over time based on each date.

8. The system of claim 7, wherein the operation of generating a sentiment value from the sentiment values includes determining at least one sentiment for the at least one feature based on the text analysis.

9. The system of claim 7, wherein the product is a service.

10. The system of claim 7, wherein the operations further comprise:

identifying an audio communication; and

converting the audio communication to text.

11. The system of claim 7, wherein each date is selected from:

purchase date; and

a communication date.

12. The system of claim 8, wherein the operation of generating the sentiment value includes combining a plurality of sentiments for the at least one feature.

13. A computer program product for reporting sentiment of a product, the computer program product comprising:

a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code, executable by at least one processor of a computer is configured to perform;

performing a text analysis on communications;

determining at least one feature for the product based on the text analysis;

generating the sentiment values using the communications for the at least one feature for the product based on a sentiment dictionary and sentiment rules that determine a sentiment strength;

determining a date associated with each of the sentiment values by extracting the date from the communications;

for each date associated with each of the sentiment values, recording a feature annotation, a sentiment annotation, the sentiment value, metadata, and the date, wherein the feature annotation is generated using a feature dictionary and feature rules, and wherein the sentiment annotation is generated using the sentiment dictionary and the sentiment rules; and

reporting how the sentiment values changed over time based on each date.

14. The computer program product of claim 13, wherein the computer readable program code generating a sentiment value from the sentiment values is configured to include determining at least one sentiment for the at least one feature based on the text analysis.

15. The computer program product of claim 13, wherein the product is a service.

16. The computer program product of claim 13, wherein the computer readable program code, executable by the at least one processor of the computer, is configured to perform:

identifying an audio communication; and

converting the audio communication to text.

17. The computer program product of claim 13, wherein each date is selected from:

a purchase date; and

a communication date.

18. The computer program product of claim 14, wherein the computer readable program code generating the sentiment value is configured to include combining a plurality of sentiments for the at least one feature.