SYSTEM AND METHOD FOR RETRIEVING AND PRESENTING CONCEPT CENTRIC INFORMATION IN SOCIAL MEDIA NETWORKS

• US 20140337328A1
• Filed: 05/07/2014
• Published: 11/13/2014
• Est. Priority Date: 05/09/2013
• Status: Abandoned Application

• Current Assignee: Veooz Labs Private Limited
• Original Assignee: Veooz Labs Private Limited
• Inventors: SARVABHOTLA, KIRAN, VADDEPALLY, SRIKANTH REDDY, RAO, PINGALI VENKATA VARA PRASAD
• Application Number:
  US14/271,504
• Publication Number:
  US 20140337328A1
• Time in Patent Office: Days
• US Class Current: 707/723
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

The embodiments herein provide a system and method for retrieving and presenting concept centric information in social media network that allows a user to search, read, express, and debate on opinions on a particular concept. The system comprises an input module for receiving an input query, a visualization module for visualizing the retrieved concept centric opinions, a topic mining module for mining a semantically related topics, a tweets tracking module for tracking influential tweets, an interesting comments searching module for searching comments posted by interested users from social media networks, public forums, blogs and other community portals, a posted comments counting module for calculating a total number of posts processed, a comments posting module for allowing users to post comments, a BUZZ words display module for displaying context words, and a new posts display module for displaying new post related to the user input query.

Claims

1. A computer-implemented system for retrieving and presenting concept centric information mined from an online portal, the system comprising:

an input module for receiving an input query from a user for retrieving a plurality of concept centric information;

a visualization module for visualizing the retrieved concept centric opinions;

a topic mining module for mining a plurality of semantically related topics for the user input query based on prior knowledge about the input query and contextual information, and wherein the contextual information comprises one or more concept centric information;

an influential comments module for analyzing and presenting one or more comments retrieved from one or more influential peoples and one or more experts in the areas related to the user input query;

an informative comments module for analyzing and presenting a plurality of comments which are grammatically well-formed, descriptive and hence readable from the retrieved concept centric information, and wherein the informative comments module comprises an informative score algorithm for calculating a relative entropy of a given comment with respect to a more random and general language model and prioritizing high quality informative readable comments based on the score;

a posts count module for calculating a total number of posts processed from social media networks, public forums, blogs and other community portals with respect to the user input query;

a comments posting module for allowing one or more users to post one or more comments to one or more social media networks directly from a concept centric social media system;

a BUZZ words module for analyzing one or more concepts associated with the concept extracted from the user input query, and wherein the BUZZ words module presents the analyzed one or more concepts in a way to show the relative significance of each association, and wherein the significance of the association is mined from the concept centric information; and

a new posts/related posts display module for displaying one or more new or recent post related to the user input query.

2. The system according to claim 1, wherein the BUZZ words module presents the analyzed one or more concepts in any of a presentation techniques selected from the group comprising tag cloud, heat maps bar charts, and pie charts.

3. The system according to claim 1, wherein the topic mining module mines the information from the plurality of online portals, and wherein the information comprising publicly posted users comment, opinions, expressions and conversations between any to users, and wherein the online portals comprises social media networks selected from a group consisting of facebook, twitter, and pinterest, discussion forums and comments in news sites, blogs, and consumer forums, and wherein the concepts can be people, places, brands, events, hash tags and any topic of discussion in online conversations.

4. The system according to claim 1, wherein the visualization module comprises:

a processor for capitalizing the concept centric opinions captured from an evidence available in the one or more posts using a classification technique and processing a number of posts updated in the social media networks and wherein the capitalizations of the concept centric opinions are mined from the plurality of posts on the internet, and the frequency of occurrence of the posts is calculated;

a display module for exhibiting the concept centric opinions for which a sentimental analysis is performed and displaying updated posts in the social media network continuously as a 24 hour activity;

a concept name display module for analyzing plurality of concept centric comments for finding the generally used display name for the concept;

a live sentiment tracking and analyzing module for displaying the concept centric opinions and calculating the percentage of posts containing positive and negative opinions;

a live sentiment display module for presenting the positive and negative sentiment as a single digit comprising a mix of mathematical symbols and colors to distinguish between a positive, a negative and a neutral sentiment;

a trend graph module for creating a trend graph based on a live sentiment with respect to time period;

a share button for allowing user to share a snapshot of the sentiment card comprising the concept name, current live sentiment and trend graph within the social media networks and wherein the social media networks comprise Twitter™

, Facebook™

, and pinterest; and

a follow button for allowing user to follow a concept centric opinion and receive updates on user e-mail periodically at preset time intervals, and wherein the updates comprise the live sentiment snapshot at a time of sending the update and the trend graph of the live sentiment;

5. A computer-implemented method executed on a computing device for retrieving, presenting and posting concept centric opinions in a social media network, the method comprising steps of:

receiving an input query from an user for retrieving a plurality of concept centric opinions using an input module;

visualizing the retrieved concept centric opinions using a visualization module;

mining semantically related topics to the user query based on the prior knowledge about the input query and analyzing the retrieved concept centric information;

analyzing and presenting one or more influential and expert comments for the user query using the computing device;

analyzing and presenting a plurality of informative comments for the user query using an informative score algorithm in the computing device, and wherein the informative score algorithm calculates a relative entropy of a given comment with respect to a random and general language model and prioritizes comments based on the score;

calculating a total number of posts processed from social media networks, public forums, blogs and other community portals correspond to the user input query by using a posts counting module;

allowing one or more users to post one or more comments to one or more social networks directly from the social media networks using a comments posting module;

displaying one or more context words of the user query, and wherein the context words of the user query are calculated using a frequency of occurrence of the keywords in the plurality of posts retrieved from the social media networks using a BUZZ words display module; and

displaying one or more new or recent post related to the user query using a new posts/related posts display module.

6. The method according to claim 5, wherein the semantically related topics are mined and retrieved from the one or more posts, and wherein the retrieved semantically related topics are displayed along with a live sentiment and a trend graph thereby allowing the user to explore other semantically related topics.

7. The method according to claim 5, wherein the comments posted by the interested users in the social media networks, public forums, blogs and other community portals are searched for the user input query, and wherein a reading of a text of the posted community is also considered in calculating the live sentiments of a comment.

8. The method according to claim 5, wherein the total number of post counts is displayed along with a source wise break-up thereby allowing the user to get an overview of an amount of activity occurred on the topic in the plurality of social media networks.

9. The method according to claim 5, wherein the context words of the user input query are calculated using a frequencies of occurrence of the keywords in the plurality of posts that are retrieved from the social media networks, and wherein the context words are displayed in either a higher or a lower font respectively to indicate a degree of occurrence of a BUZZ word in the context with respect to a given topic, and wherein the user is allowed to filter the posts based on the displayed BUZZ words.

10. The method according to claim 5, wherein the informative score algorithm calculates the relative entropy of the given post and assigns a score for the comments posted in the social media networks, public forums, blogs and other community portals, and wherein the comments searching module ranks the content items in the given context based on the assigned scores.

11. The method according to claim 5, further comprises a gender assessment algorithm for detecting a gender of the user based on a first and last name of the user.

12. The method according to claim 5, further comprises an age estimation algorithm for automatically detecting an age-group of the user based on a language used by the user.

13. The method according to claim 5, further comprises a location detection algorithm for automatically detecting a location of the user based on the posted comments.

14. The method according to claim 5, further comprises a user interest estimation algorithm for automatically detecting a user-interest based on the posted comments.

15. The method according to claim 5, further comprises a contents linking algorithm for linking a plurality of contents and wherein the plurality of contents includes articles, blogs, website links, pictures, videos, posts and comments with each other for providing a concept centric opinion.