Dawning Big Data application: Excavating the massive data gold mine in the broadcasting industry

As a long history of development and relatively mature ecosystem, radio and television industry plays an important role in the process of China's modernization construction. However, with the rapid development of the communication carrier and the content of the Internet, especially the social network of micro-blog and micro-credit and the new media, the impact on the traditional radio and TV system is more and more obvious. So, in the context of the big Data era, how to deal with radio and television? How to choose a suitable platform? How to tap and use large data technology to achieve "network linkage" to provide more quality services?

To this end, dawn with a solid large data technology accumulation as the foundation, launched a large data solution for the radio and television industry, can help the radio and television industry efficient analysis of mass dissemination of data, in-depth excavation of user needs, in the era of large data to obtain a more powerful forward momentum.

From "Mass coverage" to "public interaction" and the integration of radio and TV industry

The practice of the American Play "card house" proves that the new media in the large data information high-speed circulation platform, as the television content production service object, the audience receives the dissemination content to produce the massive interactive content, like the program impression, the comment, even the self-made content text and so on, this information quickly gathers and transmits to the television broadcast platform, This can not only enrich the interaction of content products, enhance the enthusiasm of the audience, but also the traditional television content production methods gain. But the radio and television industry is different from the new media in the large data technology application model, in the push needs more accurate analysis of the structure of the needs of this group, and no longer like the previous ratings survey to adopt the method of sample households.

In the new media era, the user's sense of hunger is increasing, but the tolerance of unrelated information is decreasing with the day, which requires the radio and television industry to use large data mining user needs, to provide users with truly valuable information.

In addition, from the individual to the group (such as family) changes, not only the number of information consumers change, consumption patterns and characteristics have also changed. In front of the TV screen, user behavior is no longer a simple fragmentation and central, and there is a group of choice offset, that is, large screen before the user behavior has a choice of accumulation trend, but the accumulation of personalized. So, to grasp the personalized trend and the combination of the characteristics of the group needs, we must rely on large data technology.

Focusing on the characteristics of mass data in radio and television industry based on high performance computing

In the great environment of the great data age, our country's radio and television industry is adapting itself, it is both the producer and the owner of the information data, and it needs to become the users and beneficiaries of the information data gradually.

To this end, dawn for many years in the field of high-end computing solid technology accumulation, Shuguang company in the country carried out a large data era strategic layout. In recent years, Dawn is based on innovative ideas, in the original mass data processing platform to integrate the latest Hadoop distributed framework, so that it can more efficient processing of structured and unstructured data. Dawning of the large data integration solution is the adoption of innovative data processing technology and open architecture, can be for large data applications to build a truly efficient, intelligent, easy-to-use holistic solution. In the application experience, it can help customers fast and intelligent analysis of massive data to improve insight and make informed decisions, and for customers from a large number of data mining the value of the "gold mine."

Large data accurate evaluation "relationship" 7 modules win in the media age

At present, the dawning large data solution has taken the lead in the financial, telecommunications, education, national defense and other fields to achieve the successful application of large data, and for large data in the radio and television industry, the potential application of a large data-broadcasting industry to tailor the solution.

The program (using Shuguang xdata-hadoop optimization software) can be divided into the following 7 large modules: Data acquisition module, calculation module, data storage module, application analysis and display module, System Monitoring management module, large Data warehouse and backbone interconnect module, including:

L HDFS main Storage module and MapReduce-oriented computing module only adopts a pass-through method, other modules of data and management communication through the backbone of the interconnected platform.

L Data Source acquisition module mainly contains three kinds of data types: Sample Set-top box data and VOD video capture data, set-top box operation log data collection and analysis, collect strategy to control collection.

L Data Computing module is mainly responsible for all the data ETL, clear, encapsulation and loading of services, while the analysis of the implementation of the task in this module.

L Application Analysis Display module is mainly responsible for b/s data analysis results display, and related data application deployment.

L system monitoring and management module is responsible for monitoring and managing the software and hardware components of the platform.

This solution builds large data platform based on server cluster, can improve the existing unstructured data off-line service level, breakthrough in the traditional technology means can not solve the massive data storage elastic expansion ability, backup recovery difficult problem; using structured and unstructured historical data resources to centralize advantages, to provide business with more integrated, Longer time span, more dimensions of data query services, improve the query access efficiency of related services, improve customer experience, so that offline data truly online and fully utilized to improve the use of a large number of semi-structured unstructured data utilization, to provide the basis for business development.

Broadcasting industry presents "two big" application trend big data becomes the marketing sharp weapon

In the light of the Dawn, the 2014 broadcasting industry large data application trend can be divided into the following two categories:

The first category: Ratings forecasts and program marketing

There is a positive relationship between the ratings of TV programs and the amount of tweets discussed. The amount of micro-blogging discussion can be an important indicator for predicting TV ratings. At the same time, for the program promotion, the user who participates in the topic discussion of the program, is not the television audience, is the network video audience. These user IDs are the audience for this program. If you put ads or push content against these users, it will be highly efficient.

The second type: Platform network linkage of Cross-platform marketing

With the complementary value of TV and network platform becoming more and more prominent, "network linkage" has become an important feature of Internet media and traditional TV media integration. Before, Weibo has been trying to network linkage marketing solutions. Specifically, Weibo will be the hot TV program set as a "Hot topic", in the PC end and mobile end set topic entrance, attract users to enter the "Hot Topic page", causing netizens to share and interact, further enlarge the name of the TV business marketing rights;

In addition, while the program is live, micro-blog will be synchronized in the information flow on the topic of the program to recommend, to ensure that the time period to show the opportunity to enhance the effectiveness of the enterprise again; through the user to obtain information in the core area-flow and topic list of recommendations, the topic of the maximum heat, so as to enhance the brand awareness

Therefore, we can foresee 2014 years of broadcasting industry, large data applications will be more targeted, users would like to be in decision-making power, insight discovery and process optimization ability to be helped. To this end, the dawn in 2014, will continue adhering to the "platform, deep application" of the big Data strategy, continue to increase investment in research and development, in the existing Parastor cloud storage, Cloudview Cloud operating system, xdata large data platform, forming the core technology group, to create a better product. At the same time, the dawn will take the lead in various industries to form a large data application of the best practices in the radio and television data industry applications plowing.