Win-win rule: the perfect combination of video sites and big data

At first, big data refers to a dataset that is too large to be statistically and analytically analyzed, but has since evolved into a way of analyzing massive amounts of data and obtaining the services of great value that we need .

This is an increasingly valued form of analysis of these data is difficult to store these data as difficult as possible, but also need to use some of the unprecedented way, most of the country is now using Hadoop open source distributed data cluster to meet the The demand for data statistics.

Many sites have big data strategies, especially in the e-commerce and video industries, because of the subtle data that can be analyzed in huge amounts of data with great value benefits.

But more challenging and representative should be for large data mining video sites, after all, e-commerce for big data, there are still many advantages, such as e-commerce users choose anyway, and ultimately will select a product , And then register, log in, pay, virtually all of their most important information to give out. However, as the vast majority of video site users are not registered, how to record such information in the case of how to do big data analysis?

A recent Netflix home-made play, The House of Cards, has drawn everyone's attention and is said to be a key element of Netflix's deduction from the data for the success of House of Cards: users who love BBC episodes, the performance of David Finch, · Shi Pai Xi's performance portrayed. Eventually, Netflix was a huge success with the conclusion that based on Big Data was very reliable, "House of Cards" caused a dramatic boom in the United States and more than 40 other countries.

Big data who do the best?

So we have a question, some people in the video industry is big data? Who do the best?

Is Netflix? Technically, I do not think so.

Netflix is ​​still the success of its data can be tapped, Netflix is ​​an online DVD rental site, through direct rental data users can know that users like the BBC's episode, like the film director David Finch As an alternative expression of social issues, and Kevin Staples in the defacement of all kinds of sinister cunning villain that superb acting, so successfully launched the "House of cards," this high-profile US drama.

And in the anonymous user big data is the best than Youtube, after I came into contact with this amazing video site, not only impressed by his precise recommendation algorithm, but also for their admiration for the openness of the data, as long as you want Statistics and analysis, Youtube will be available to almost all video data to you, thanks to their strong support of Google technical support.

All of Youtube's big data messages are organized around a single video program. The first visit, the first share, the first comment are all displayed in front of you, including the time and URL listed, which is A kind of open mind.

There are also several domestic video sites for video data bigger: Youku, Sohu (full data); Cool 6 network, potatoes, 56 (to provide some data); Fantastic Art, Music, Tencent (statistics but not provided); They also will be the video site big data contender.

Localization of big data applications

The benefits of big data analysis is responsive, on-demand access; for the domestic video site profit model, what areas can produce value?

Content recommendations regarded as the most basic application, to provide users with more appetite video, increase click-through rate, get more advertising exposure, is the most direct form of profit. This is also the easiest and easiest to implement, and is also commonly used by major video sites, but it depends on the accuracy of the algorithm. Many sites simply use the words, tags and other forms of recommendation for the user video that is accurately recommended, and its title "recommended algorithm" four words, we can not talk about what, after all, based on big data recommendation algorithm is not Simple project.

Followed by the data will be filtered according to the conditions of users, to promote the profitability of content in different dimensions; such as film and television works of network promotion, content preheating, pay-video recommended on-demand. Subdivision through the user base, resulting in a higher conversion effect; but for its supporting basic services are more numerous. Such as membership system, payment system, billing system, video content encryption, and fuzzy recommended functions, will take some time to develop, the relative need to invest more manpower and development costs.

Copiable business model?

Are there any other industries based on big data profit model has been verified, and the model can be copied, applied to the video site?

Netflix has provided us with a great example of how to select the director and actor by analyzing the user's preferences, which can be perfectly reproduced in the country. 2013 Youku, Fantastic Art are betting on homemade drama. In the first two years, there were also several successful micro-movie series in China, such as "Adolescence" series, and later several more popular "godfather", "host" and "mommy" mostly won the eye-catching way User attention. If big data analysis can be successfully filmed to achieve the quality and taste of radio and television standards to achieve the home-made drama in China also create a successful case, it will be possible to lay its hegemony.

Another example is to copy Baidu's auction advertising model, by providing users and advertisers marketing big data analysis support, allowing them to run their own ads; let advertisers find valuable videos, ads, generate a self-marketing The ecological environment will benefit the development of websites and industries.

Valuable video has no advertising value, valuable video by advertisers have been bidding for competition, through the video itself to improve the original author's economic income, can be described as a win-win principle; not only to encourage the enthusiasm of the original creator, you can Let the quality of the site's video content continue to improve, the formation of difficult to exceed the value of height.