The trial operation of big data system for comprehensive evaluation of radio and television program ratings will solve the problem of ratings fraud.
Cctv news(Reporter Tang Shiying) On December 26, the State Administration of Radio and Television held a press conference in Beijing, announcing that the big data system for comprehensive evaluation of radio and television program ratings of the State Administration of Radio and Television was basically completed and put into trial operation.
Press conference site
At the press conference, Yu Ying, president of the Radio and Television Planning Institute of the State Administration of Radio and Television, introduced the relevant situation.
China’s ratings survey began in the mid-1980s, mainly using corresponding technical means and methods, such as using recorders and sampling surveys, to investigate, count and analyze the audience’s viewing of programs. Relevant data can be used as a reference to measure the copyright value of corresponding programs and the advertising value of corresponding time periods. With the rapid development of information network technology, the increasingly fierce media competition and the deepening of industry reform and innovation, audience survey has attracted more and more attention from all sides. At the same time, it also puts forward new and higher requirements for further improving the methods and means of audience survey by using new technologies and effectively enhancing the scientificity, timeliness, security and authority of audience survey.
The big data system for comprehensive evaluation of radio and television program viewing, which was put into trial operation this time, mainly collects the viewing behavior data of a large sample of users by establishing a secure channel with network transmission institutions, and outputs 30 core indicators such as the number of users on, users watching, ratings and market share after cleaning, conversion, analysis and mining. The system has the following obvious advantages:
First, there are many samples, wide coverage, and super-large information sources. At the beginning, the system collects the viewing data of 40 million cable TV and IPTV sample users, covering a variety of viewing methods such as live broadcast, review and on-demand, and will gradually expand to hundreds of millions of sample sizes to achieve full coverage of samples.
The second is big data and cloud computing, and real-time processing is accurate to the home. It can not only reflect the viewing situation of popular programs and prime time, but also accurately capture the viewing characteristics of minority programs and marginal time. For example, if 500 sample users are selected from 5 million users in a city, the programs that these 500 sample users have not seen will be counted as "zero viewing"; But in the large sample statistics of the system, even if 1‰ The low ratings will also leave a mark.
The third is to prevent manipulation and pollution, and solve the problem of viewing fraud. The data collection, cleaning, analysis and presentation of the system are seamlessly connected, and the whole process is automated and closed, which can effectively prevent human manipulation. Based on massive data statistics, the influence of individual sample data pollution on statistical results can be ignored. Similarly, for a city with 5 million users, if the user data of 5 out of 500 samples are polluted for a program, the audience rating of the program can be increased or decreased by 1%. If the real audience rating of the program is 2%, the statistical deviation can reach 50%. For this system, this situation will only affect the ratings of the program by one millionth, and the statistical error is basically negligible.
Fourth, multi-dimensional, all-round, comprehensive evaluation leads development. The system can not only provide objective and true viewing statistics, but also build an all-round comprehensive evaluation model of programs based on multi-dimensional modeling and analysis of program communication, guidance, influence and credibility, combined with ideological, innovative and professional program quality elements. Through deep mining and timely feedback of viewing data, it can also be used to guide content selection, material integration, demand combination, analysis and prediction, creative production, change the traditional program production mode, and promote the high-quality development of radio and television content with big data.
The fifth is all-media, openness and a new orientation for the future. The system actively adapts to the development trend of technology and communication, and will comprehensively cover different communication channels such as cable TV, direct broadcast satellite, IPTV, Internet TV and network audio-visual field, and preset the new positioning and model under the general trend of national cable TV network integration and 5G mobile application in advance.
State Administration of Radio and Television, Central Radio and Television Station, National Bureau of Statistics, many provincial cable TV network companies, IPTV transmission service enterprises, radio and television stations and other relevant departments and experts from some scientific research institutes and universities attended the press conference.