Design And Development Of Audience Rating Analysis And Forecast System

In recent years, with TV media continued to expand the ratings are statistical datahas become the TV media to evaluate its programs this year, to determine the variouschannels and air time advertising prices and advertising customers to choose the media,channel, time, and an important basis for program types, however, the current televisionratings survey is far from perfect, one of the ratings survey data is relatively simple andcan not affect program ratings, quantitative analysis of many factors; the other is thecurrent ratings system is still only able to show the investigation carried out broadcastof the investigation, analysis and evaluation, but can not be broadcast before thepredicted ratings. Papers on the ratings of the factors studied, using J2EE technology toachieve the ratings analysis and forecast system for the ratings for broadcast before theprediction.Ratings in the current survey system, the television media is basically to use theexisting maximum ratings research company Nielsen ratings survey for the survey datais Nielsen data as a television program ratings data. With the increasing size of thedatabase, the data volume of up to several GB or even TB level, these data contain agreat deal of valuable information is not used, in order to more effectively manage andutilize the data in the database, access to knowledge and values in which the potentialThis paper use data mining modeling idea for the massive historical time Nielsen ratingson the test data to establish a simplified data warehouse and use the data mininganalysis of the classification and prediction methods, the factors that affect the ratingsCluster analysis, access to affect the ratings of the relationship between factors.Including the media, the value of broadcast time slots, channel effectiveness,programming brand, and the formation of a quantitative evaluation index system andaccurate performance evaluation criteria, and thus on the future of TV programming andadvertising value of measurable ratings intuitive forecast leadership for the televisionmedia, brand development for the program choice and the decision to provide detailedand reliable quantitative basis for guidance.Effectiveness of the program time, channel evaluation, program type, channel coverage, actors, directors, and other factors affecting ratings, statistics and quantitativeanalysis and reasonable in a specified period of time the evaluation index, andrespectively, intuitive, user-defined charts shows that for example a zoom of the3Dcolumn chart, pie chart, line charts and so on. In terms of viewership forecast, mainlywith the principal component analysis method, weakening remove merge some of theless impact on the relationship between factors that influence the weights to generateone-sheet, further ratings, indices and quantitative relationship between thesefactors-type and ultimately the impact factor according to their weight and therelationship between the proportion of organic combination of options will be based onuser demand for listing of programs broadcast channel, program type, playing time,covering the network, actors, directors, chart type, etc., according to relationship thecalculation of the program in the specified period of time the ratings charts, which willbe broadcast on the future of television program ratings for sound scientific quantitativeprediction.