《新的电视节目的观众评价外文翻译.doc》由会员分享，可在线阅读，更多相关《新的电视节目的观众评价外文翻译.doc（7页珍藏版）》请在三一文库上搜索。
1、毕业设计（论文）译文及原稿译文题目 新的电视节目的观众评价基于GM（1.1）包络模型预测 原稿题目 Audience Rating Prediction of New TV Programs Based on GM (1.1) Envelopment Model 原稿出处 Control and Decision Conference (CCDC), 2010 Audience Rating Prediction of New TV Programs Based on GM (1.1) Envelopment ModelZheng YileiAbstract Audience rating i
2、s the core index of TV programs as well as the embodiment of core competitiveness of TV stations. Daily management of TV programs is mainly audience rating-centered. However , posterior survey and evaluation concerning audience rating has been incapable of meeting the demands of daily management of
3、TV stations. It is then an important task for modern TV stations to apply scientific and effective approaches to predict and analyze the audience rating of programs so well as carry out pre-experience control in daily management. On the basis of comprehensive knowledge of the current situation of ne
4、w TV programs audience rating management, this paper analyzes the significance of Audience rating prediction of new TV programs in detail;based on the small data quantity and great undulatory property of new programs audience rating, construct small sample and few information-based GM (1.1) envelopm
5、ent model to predict and analyze it; in the end, great effect is achieved by verifying the result with some casse study. This paper provides a new method and insight for oscillation series prediction studies under small samples and a few information. In order to explore the competition dynamics and
6、determine how they affect the audience rating of TV stations, a dynamical model of TV stations is proposed and discussed in this paper. This model shows the well known appearance of winner-take-all phenomenon among the fierce competition from many TV stations. And then, dynamics competition analysis
7、 between two TV stations is discussed in detailed. It is found that, the famous winner-take-all phenomenon also exists among TV stations competition processes, especially among the stations which serve almost the same types of programs. Whats more, this paper proposes some strategies for those stati
8、ons, which is helpful for the station with a large initial audience rating and proper reform period. So, to be the winner, good propaganda and reforms are both critical and necessary. Lastly, Numerical simulations further proves the effectiveness of the analysis.I. FOREWORDUDIENCE rating survey is a
9、 subject of study conducted by professionas on the effect of the broadcast of programs taking into the consideration of the status quo of the rapid development of TV industry . Audience rating is an important index measuring the quality of TV programs, but the audience rating survey and evaluation a
10、fter the broadcast of TV programs have been incapable of meeting the demands of various TV stations that are under ever-increasing fierce competitions. Therefore, scientific and effective method for audience rating prediction is in dire need. On the in-depth study on GM (1.1) envelopment model and a
11、pplication of data mining predictive model set up hereby to the audience rating prediction of TV programs, this study shall help put TV programs productionin a more scientific and digital direction.II. SIGNIFICANCE OF AUDIENCE RATING PREDICTION OF NEWTV PROGRAMSA. Importance of Audience rating No on
12、e can doubt the importance of audience rating survey on the present TV media. The statistical data of audience rating has become the core basis for TV media to evaluate its own programs and determine the program arrangement at each channel and time interval; meanwhile, the advertising Y. Zhang is wi
13、th the College of Econcomics and Management, Nanjing University of Aeronautics and Astronautics, Nanjing, Jiangsu ,China. (Phone:008613505155550;e-mail:). department of TV Station works up the marketing strategy of advertising programs and prices through audience rating and again, the advertising ag
14、ency and advertiser also evaluate the rationality of advertising price fully by the advertising cost deducted from Audience rating. Hence, Audience rating also becomes the fundamental of advertising business of TV stations. B. New TV programs are means for TV station to gain victory in market compet
15、itions The importance of TV programs to TV stations just as that of the products from the production line to the enterprise. Good programs and programs with high Audience rating enable TV station to establish great advantage and its leadership position, particularly in a “remote era”, when the targe
16、t viewers hold the right “to watch” or “not to watch”, the pressure of TV stations on TV programs is unprecedented. Such product development strategies as leader strategy, follower strategy, dark horse strategy and market nichestrategy, etc. which are quite familiar to Chinese enterprises can be per
17、fectly applied to the discussion on the special product development of TV programs as well. A new program with clear positioning, reasonable arrangement, novel content and fine packing shall be the major work for the innovative product development of TV stations, shall be an important content in the
18、 macro decision-making of TV stations, shall be means for TV stations to gain victory in market competitions and shall be the foundation for the survival and development of TV stations even more. Latest survey indicates that the life span of a new TV program product shall not exceed two years. While
19、 the program is aging, measures such as layout change and program product innovation must be adopted to upgrade the Audience rating of TV programs so as to adapt to demand of ever-increasing fierce competition. C. Great significance of audience rating prediction of new TV programs For the time being
20、, Audience rating survey system is still confined to survey, analysis and evaluation of programs after its broadcast and is incapable of audience rating prediction before broadcast. As a result, adjusting program layout and launching new program are more often qualitative analysis by personal experi
21、ence rather than scientific quantitative study. How to reduce errors arisen during program layout adjustment and new program broadcast? Scientific prediction and analysis to the future Audience rating in the initial stage after broadcasting the program shall be made to obtain convincing prediction o
22、f audience rating data. Audience Rating Prediction of New TV Programs Based on GM (1.1) Envelopment Model Zheng Yilei AAudience rating prediction is a dynamic process. So, dynamic analysis shall be made to the Audience rating prediction and factors influencing Audience rating shall be aken into full
23、 consideration to predict the time schedule while the program entering each developmental stage. For instance, when shall a newly broadcasted program enter maturing stage and whats its expected Audience rating. When will it decline? To what degree of the Audience rating drop can be regarded as a sig
24、n that the program has entered decline stage and how to extend the life span of this program and the like. A good program may come to an untimely end, if adjustment has been made to the program having not entered the maturing stage yet because the audience rating fails to reach the set highest targe
25、t value,. Audience rating prediction is research before broadcasting and audience rating evaluation is research after broadcasting, this is a linked circulatory system. Only via study of this system, the more scientific basis can be provided for the launching of new programs and program layout. Sett
26、ing up a audience rating survey system which combines prediction with evaluation enables each key point to be placed at a level of scientific investigation, avoids negative effect brought to the program planning due to deficiency and difference of personal experience. It plays a positive role in the
27、 development of the whole TV industry. In fact, this mechanism can also be extended and used in some competition behaviors analysis in other related fields.III. CONSTRUCTION AND SOLVING OF GM(1,1) ENVELOPMENTMODELSince the broadcast time of new TV programs is short and the data accumulation is insuf
28、ficient, the audience rating of these new programs displays such features as few data, few information and uncertainty. Therefore, traditional statistical and quantitative methods can not be used to conduct modeling research on audience rating, and the prediction precision is also difficult to meet
29、the demand. Concerning that , this paper prepares to conduct prediction research on audience rating prediction of new TV programs under little data and few information by virtue of the GM (1,1) model of gray system.However, since the influencing factors on audience rating of TV programs are various
30、and complicated; the data of audience rating is very disordered, the undulatory property is quite large, and the index property is not strong, simple application of the classical GM (1,1) model will have insufficient simulation accuracy and subject to big errors. In consideration of the above reason
31、s, data envelopment principle is combined with the classical GM (1,1) model to construct the GM (1,1) envelopment model, treating the original data with strong undulatory property so as to improve prediction precision and reduce prediction error. The principle and algorithm of GM (1,1) envelopment m
32、odel are as follows:STEP1 Assume X (0) as original data sequence. Construct lower edge envelop curve ( ) u f t and upper edge envelop curve ( ) s f t for X (0) according to data envelopment principle, and get lower edge sequence (0) u X and upper edge sequence (0) s X of X (0) according to ( ) u f t
33、 and ( ) s f t . STEP2Carry out data accumulation treatment on (0) u X and (0) s X to get 1 AGO sequence of (1) u X and (1) s X . STEP3Calculate smooth degree (k) and class ratio (k) respectively in accordance with date sequence or not.If both are satisfied, GM (1,1) modeling operation will be condu
34、cted; otherwise, go back to the last step and carry targeted inspection on failed sequences, make accumulation operation again until the requirements are satisfied.STEP4 Construct GM (1,1) models for qualified envelopment sequences of (1) u X and (1) s X respectively and make prediction, through whi
35、ch to get the minimum prediction sequence (0) ( ) u X k and the maximum prediction sequence (0) ( ) s X k . STEP5 Determine experts coefficient of optimism (0 1) on this program with expert opinion method. Conduct weighted processing on the minimum prediction sequence (0) ( ) s X k and the maximum p
36、rediction sequence (0) ( ) s X k and get the prediction sequence (0) ( ) (0) ( ) (1 ) (0) ( ) s u X k X k IV. CASE STUDYWe chose the audience information collected by city network with regard to a new TV programFeichang Bu 1 Ban of XX TV station from the beginning of 2009 to the early March. In orde
37、r to guarantee the rationality of data and eliminate partial time error, this data is the mean value of the 7-day audience rating in a week. The original data is seen in Table 1, and the data broken line graph is seen in Fig. 2.V. CONCLUSION.The actual audience values of the following three phases o
38、f the new TV programFeichang Bu 1 Ban of XX TV station are respectively 3.82, 3.85 and 4.39, which has a prediction error of less than 5%. The empirical study of this paper shows that: according to such features as small audience rating data and large undulatory property of the new program, the cons
39、truction of GM (1,1) envelopment model on the basis of small sample and few information as well as the prediction on it have achieved good effect. This paper has provided a new method and thought for prediction and research of oscillation sequence with small sample and few information; besides, itAs
40、sume X (0) as original data sequence Construct lower and upper edge envelop curve ( ) u f t ( ) s f t Get lower and upper edge sequence (0) u X (0) s X Pass the test of smooth degree and class ratio Carry out sequence accumulation treatment No Construct GM (1.1) models, get the (0) ( ) u X k (0) ( ) s X k Conduct weighted processing ( ) (0) ( ) (1 ) (0) ( ) s u X k X k X k以下是说明文字，正式成文后请删除。外文原稿可以直接使用复印件。