BrandTracker: Tracking Brands in Social Media (13224N)
IDA Technology Roadmap 2012
This technology falls in the following categories of Singapore's IDA Infocomm Technology Roadmap 2012:
This invention is a brand tracking method on social media platforms. More particularly, it is a method and system that uses the social context and the image information of microblogs to filter tweets related to a given brand in order to rank them and produce a more accurate product tracking functionality. This technology could be used, for instance, by companies interested in monitoring and analyzing their public impact in social media.
The invention's mechanism consists of three key procedures: 1) seed tweets generation via logo detection and image relevance analysis; 2) extended search by social context, and 3) image similarity and a joint text-visual learning-based method for ranking.
For more information on this technology contact:
The invention has been tested experimentally on the Sina Weibo platform, the largest microblog platform in China. Evaluation results indicate that, in comparison to existing methods, this invention is able to significantly improve brand tracking performance in social media platforms.
A demo of this technology can be found here.
Technology Readiness Level 5 on the scale by the Ministry of Defence Singapore.
About the Research Group
Professor Chua Tat-Seng is founding Dean of the School of Computing at NUS and director of the NExT Center, a joint venture between NUS and Tsinghua University with the support of MDA of Singapore. The center focuses on crawling and mining user-generated content in Singapore and China in the areas of location-oriented information, topic-oriented information, application-oriented information and structured information.