Cozhou joins hands with Guiguang Network to apply radio and television big data
The massive data generated during the content production and dissemination of the broadcasting industry contains rich value. Exploring and practicing the application of big data technology in the broadcasting and TV industry will help improve the capacity of broadcasting and TV content production, innovate broadcasting and TV business products, and improve broadcasting and TV operation services Therefore, it will help the integrated development of radio and television media and promote the industrial upgrading of the radio and television industry.
It can be seen that there is no shortage of big data in the radio and television industry. The key lies in how to apply these big data! Co-Continent and Guiguang Network have made the best comment on this issue on the road to the practice and application of radio and television big data.
As of November 2019, the big data system of Guiguang Network has collected more than 3.1 million set-top boxes, and generates more than 1.5 billion raw data every day. Through data cleaning, about 40 million complete playback records are generated every day.
In 2019, after implementing the data collection of all two-way set-top boxes, Guiguang Network began to build a data analysis platform. Through the analysis of the collected data, it realized the live broadcast ratings, on-demand timely moving back to the ranking analysis, user loss warning, network traffic monitoring warning , Special area page order rate, end-user behavior event analysis and other functions, and based on the big data system analysis results to connect to the SMP network management, realize user-on-demand error code statistics and early warning, user call ticket extraction and other functions to help operation and maintenance personnel quickly locate faults. At the same time, Guiguang Network's big data platform provides data to third-party business systems such as recommendation systems and advertising systems, enabling new services such as intelligent recommendation, precise advertising, and personalized portals to enhance user experience.
Next, we will take you to see how Cozhou has taken three steps to practice the application of radio and television big data with Guiguang Network!
1. Technology-driven, building the overall architecture of a big data platform
The key technologies of big data cover many aspects including data storage, processing, and application. According to the processing process of big data, it can be divided into data collection, data preprocessing, data storage, data analysis and mining, and data application. Therefore, in terms of platform architecture, the main architecture of Guiguang Network's big data platform is data collection layer, application layer and display layer.
The data acquisition layer mainly obtains the data required by the subsystem through various methods (including system interface, monitoring equipment, monitoring module, set-top box, etc.), and through interface adaptation, provides a general data access interface to achieve data access and other operations , Shielding the impact of data heterogeneity on the platform. At the same time, it realizes the objectification of data, and provides object-oriented data access for upper-level software modules.
The system application layer mainly conducts comprehensive calculation, statistics, and analysis on the data acquired by the data collection layer through the construction of various software modules. It has different management strategies for different management objects, and the system provides data addition, deletion, modification, and query Basic management functions.
The system display layer provides multiple presentation methods such as lists, charts, graphs, etc., to facilitate users to intuitively understand and manage services and monitor current performance. At the same time, the system supports comprehensive statistical analysis of information such as set-top boxes and performance data to help users quickly and intuitively understand the operation of platform services.
2. Application function-oriented, so as to accurately perceive user needs
When building a big data platform for Guiguang Network, Co-Continent focused on application functions, fully mining media content data and analyzing user data, so as to accurately perceive user needs and transform them into the collection and production of broadcast and television content. , Program recommendation and operation, user tracking and service, monitoring and management of public opinion, program transmission and distribution, and program monitoring and supervision and other fields of data output.
The main functions include:
Data collection function——
1) Offline data collection function. Regularly collect data from high-definition interactive platforms, IP video platforms and OTT platforms through offline FTP, collect on-demand and live broadcast data from users on the live network, clean them, and store them in an offline database.
2) Online data collection function. The nginx load address is reported online in real time through the terminal set-top box acquisition probe module, and after analysis and cleaning by the online acquisition module, it is stored in the online database.
Data analysis and report function——
1) User trend analysis function. Collect the daily activity status of all interactive users on the entire network through offline data, analyze recent new users and online user data based on regional information, and present it in reports for operational analysis
2) User churn warning function. In the case of information collection based on big data, use user playback behavior data, paid behavior data, and machine learning methods to infer user data that may be lost after a period of time, provide data support for system operation, retain users, and improve work efficiency.
3) Live channel ratings data analysis and report functions. Collect live data information of all interactive users across the network through offline data, calculate the ratings of live channels and other data, and output reports for presentation.
4) On-demand, time shift and look back data analysis and user playlist report functions. Collect all interactive user-on-demand data information of the entire network through offline data, calculate and analyze media resources on-demand ranking, column on-demand ranking, user time-shift playback and user review playback, and output detailed reports such as user playlists based on regional information
5) Multicast switching list function. For IP platform users, it also collects the multicast switching status of the users while watching the live broadcast, and outputs reports for analyzing the network status information in the terminal user area.
6) Online data analysis and report functions. According to the user behavior reported by the terminal probe, the user behavior events are classified and collected, analyzed and cleaned, and the user behavior report function is output.
7) Reports on subscription rate and on-demand conversion rate of users in the zone. According to the specific area user on-demand behavior and order information data, analyze and calculate, output the user-on-demand conversion rate and the area user’s order rate report function.
8) Flow data report function. According to the network management platform of Guiguang Network, collect flow data information and output flow data report function
9) Data acquisition and data output function. According to the output specifications of the big data system, the data can be output through online interfaces, offline FTP and other methods according to the needs of third parties for use by third-party systems.
3. Keeping integrity and innovation, step by step empower Guiguang Network
With regard to the construction strategy of Guiguang Network's big data platform, Intercontinental must clarify the demand and target on the one hand and make an overall plan. On the other hand, based on business reality, step by step the construction and improvement of the big data platform. Based on the analysis of the current needs of Guiguang Network, we will step by step to promote the construction of big data business application services and increase the business on the platform.
In the follow-up planning and construction, Intercontinental will continue to develop more application functions, continue to promote the intelligent development of Guiguang Network, comprehensively enhance the commercial value of Guiguang Network, and provide support for the upgrade of Guiguang Network industry. These include:
Data collection visual editing function——
The current data acquisition system and report output function are optimized and transformed, big data is made into visual editing, data results can be customized, all fields can be customized, customized report styles, customized export results, customized query conditions .
User portrait function——
User behavior analysis based on data collection abstracts each specific user information into tags, and uses these tags to concretize the user's image, establish a user portrait function, and provide differentiated services for specific user groups.
Rating report analysis——
It can support big data analysis of advertising operations and accurate advertising. Start-up screen advertisements, channel banner advertisements, and volume adjustment screen advertisements are the most important forms of advertising for radio and television network operators. The user behavior analysis data report can provide sufficient data support for the radio and television operators. According to the viewing rate, user viewing time and other conditions, the startup screen advertisement, channel banner advertisement and volume adjustment screen advertisement are grouped and packaged according to region, time period, and ratings. , Refined operation of advertising resources.
Intelligent recommendation, improve user ARPU value. There are more and more channels for obtaining videos, basically allowing users to obtain them at any time, any place, and on any device with a screen. Video broadcasting is becoming more and more intelligent, allowing users to get the content they want to watch in the simplest way. Video intelligent recommendation has become the standard configuration of all video service operators. How to make digital TV
Users get the best video viewing experience. In addition to pushing more high-definition video program sources that cannot be achieved by other channels, we also need to perceive users’ viewing habits, actively recommend the most likely programs to users, and allow users to watch Roam between several entertainment screens in the home. In the end, users will remain in the broadcasting network, improving the utilization rate of the broadcasting network's video resources and increasing the user's ARPU value.
Cluster analysis and trajectory analysis retain customers. Big data analysis can use cluster analysis to understand the types of programs and channels that each customer likes to watch daily, as well as the likely purchase of the latest blockbuster, so that the account manager can pick out the top 20% of users who are most likely to buy the product. Take door-to-door visits or telephone promotion to quickly and effectively conclude transactions.
Operation and maintenance application——
Effectively handle failures. The user terminal uploads its status data, management control information and other data through a two-way network. The terminal management system uses a two-way protocol to perform boot management, authentication, parameter reading and writing, log reporting, upgrade management, fault diagnosis, etc. On the basis of management operation, it realizes the visual management of the user terminal, the terminal running status can be presented, the terminal can be controlled, and the fault can be remotely diagnosed. The terminal management system can provide information on the current system network and business operation status with various granularities with customized reports, provide system managers with an understanding of the current network operation conditions, provide a basis for high-level system decision-making, and improve broadcast and television operators Service quality and management efficiency.