PROCESSING METHOD FOR SOCIAL MEDIA ISSUE AND SERVER DEVICE SUPPORTING THE SAME

• US 20150178267A1
• Filed: 04/18/2014
• Published: 06/25/2015
• Est. Priority Date: 12/19/2013
• Status: Active Grant

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: HEO, Jeong, KIM, Hyun Ki, PARK, Sang Kyu, LIM, Soo Jong, CHOI, Yoon Jae, CHOI, Mi Ran, RYU, Pum Mo, BAE, Yong Jin, YOON, Yeo Chan, JANG, Myung Gil, LEE, Chung Hee, JO, Yo Han, OH, Hyo Jung
• Granted Patent: US 9,418,058 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

The present invention relates to a social media question and answer service, and discloses a social media issue processing device and method. The method includes a question analyzing unit which analyzes a question type of an input question through a question pattern dictionary which is stored in advance; a social media analyzing unit which performs issue period recognition for the question, question type based analysis, question based summary creation, and question based reliability calculation; and a report creating unit which creates a summary in accordance with at least one of correlation between the question type and the question based summary, correlation between the issue period and the question type based analysis result, and correlation between the reliability and the question based summary.

Claims

1. A social media issue processing device, comprising:

a question analyzing unit which analyzes a question type related with a question input based on a question pattern dictionary which is stored in advance;

a social media analyzing unit which performs issue period recognition for the question, question type based analysis, question based summary creation, and question based reliability calculation; and

a report creating unit which creates a summary in accordance with at least one of correlation between the question type and the question based summary, correlation between the issue period and a question type based analysis result, and correlation between the reliability and the question based summary.

2. The device of claim 1, further comprising:

a resource DB which stores the question pattern dictionary, a natural word summary frame, and a summary report form; and

a resource building unit which analyzes at least one of the questions which are collected in advance to build at least one of the question pattern dictionary, the natural word summary frame, and the summary report form.

3. The device of claim 2, wherein the resource building unit includes:

a question type classifying unit which classifies a question type in the questions which are collected in advance; and

a question pattern building unit which detects a predetermined pattern from the classified question type.

4. The device of claim 1, wherein the resource building unit includes:

an entity type classifying unit which classifies an entity type in the questions which are collected in advance; and

an entity based social media detail module analyzing unit which analyzes a characteristic in the entity based social media for the classified entity type.

5. The device of claim 1, wherein the question analyzing unit includes:

a language analyzing module which analyzes a morpheme of the input question and recognizes an entity name;

an entity recognizing module which recognizes a major entity from the result which is analyzed by the language analyzing module based on the question pattern dictionary;

a question type clue recognizing module which recognizes a clue word of the question based on the question pattern dictionary; and

a question type determining module which determines a question type based on the number of entities and the clue word.

6. The device of claim 1, wherein the social media analyzing unit includes:

an association/competition analyzing unit which analyzes an associative word and a competitive word for the question type which is analyzed by the question analyzing unit;

an issue processing unit which performs an issue period related analysis; and

an individual social media analyzing module which compares and analyzes entities ranked by a predetermined level during the issue period.

7. The device of claim 6, wherein the issue processing unit includes:

an issue time recognizing module which recognizes an issue period based on an entity based buzz progress;

a buzz progress graph comparing module which compares issue periods which are recognized in a plurality of entities;

an issue period combining module which combines overlapping issue periods of the entities; and

an issue labeling module which creates an issue label for the issue periods of entities to be compared.

8. The device of claim 1, wherein the report creating unit includes:

a reliability based correlation analyzing module which analyzes the correlation using the reliability value;

a summary report form creating module which selects any one of summary report forms which are predefined in accordance with the question type and recreates a summary report form in accordance with the reliability value;

a natural word summary creating module which creates a natural word summary based on correlation of the individual social media analysis results which are presented in the summary report; and

a summary report creating module which creates a summary report by writing a result of the question analyzing unit, the individual social media analysis result, and the natural word summary in the summary report form.

9. The device of claim 1, further comprising:

a display module or an input module which supports an input function to request at least one of the issue period selection and the question analysis result correction.

10. A social media issue processing method, comprising:

an inputting step of receiving a question;

a step of determining a question type of the input question through a question pattern dictionary which is stored in advance;

an analyzing step of performing issue period recognition for the question, question type based analysis, question based summary creation, and question based reliability calculation; and

a summarizing step of creating a summary in accordance with at least one of the correlation between the question type and the question based summary, correlation between the issue period and a question type based analysis result, and correlation between the reliability and the question based summary.

11. The method of claim 10, wherein the inputting step is a step of receiving a natural word question.

12. The method of claim 10, wherein the analyzing step includes:

a step of creating a buzz progress graph based on the issue period recognition; and

a step of automatically comparing individual social media analysis results between entities in the buzz progress graph to understand a similar pattern.

13. The method of claim 10, wherein the summarizing step includes:

a step of presenting an insight for the correlation between the entities as a natural word summary.

14. The method of claim 12, wherein the analyzing step further includes:

a step of ranking and presenting the issue periods of individual entities;

a step of receiving an input to select a specific issue period; and

a step of presenting an insight for the correlation between the entities through social media detail analysis by a zoom-in method for an issue period selected by the input.

15. The method of claim 12, wherein the analyzing step further includes:

automatically creating and attaching an issue label for the issue period.