Big data: Can "perceive" emotions "dig" out of plagiarism

Reporter from the Shenzhen Institute of Chinese University of Hong Kong recently sponsored the "Rich Media data: awareness, processing, retrieval and Application" blog forum (phase II) to learn how to effectively mining rich media data to become a major research hotspot in the field of data, accurate modeling, design efficient algorithms, build core technology, Thus realizing http://www.aliyun.com/zixun/aggregation/13873.html "> data-driven operation is one of the main directions that industry and academia are exploring and studying at present.

"21st century's ' data ' equals 20th century's ' oil '. "According to Lu Rongcong, a professor of computer science and engineering at the Chinese University of Hong Kong and a Fellow of the International Institute of Electrical and Electronic Engineers," Unlike traditional multimedia data, rich media data are aggregated with the dynamic and interactive nature of multiple media data, including more information and value. Large data has four very typical features 4V, that is, the scale of rapid expansion (Volume), a variety of sources (produced), rapid change in Speed (Velocity), complex relationship (veracity).

In July this year, the Shenzhen Research Institute of the Chinese University of Hong Kong was awarded the key Laboratory project of Shenzhen Rich media data analysis and application. Based on the rich media large data computing platform, the core of intelligent computing technology and the development of key technologies in connection with Internet and multimedia related applications, the laboratory has accumulated abundant research results and successfully used in the Internet and education industries.

Professor of Computer Science and Engineering Department of the Chinese University of Hong Kong, Shenzhen Rich media large data analysis and Application key Laboratory director Guoqing told reporters, for example, through social network analysis can be used for opinion mining, emotional analysis, can use analytical tools to learn who to promote certain products more effective, Can be through the public social network of corporate leaders to dig out a certain chain of social relations, such as through the text of the similarity analysis, you can quickly find a suspected copy of the article, ultimately by the person to decide whether a piece of plagiarism.

Guoqing said that the current system and method of similarity of the test documents have been in Hong Kong, Taiwan, Singapore and other land applications for 5 patents, has been in the Hong Kong government and the Hong Kong universities and so on a wider range of use, hoping to do more through the Shenzhen platform to promote the use of the product in the mainland.

According to the introduction, Shenzhen rich media large data analysis and application key laboratory will not only continue to produce theoretical results, analytical tools, product applications, etc. in the rich media data, but also will rely on the advantages of Shenzhen as an industry oriented city in the large data industrialization, and apply more research results to the mainland.