Design And Implementation Of Financial Information Recommendation System For The Specified Media Type

With the frequent developing of Internet, Internet media types are more and more rich, the content is also increasing. In the era of information explosion, it is particularly urgent to find valuable information from huge data. Financial information for the specified media type recommendation system, designed to filter out financial information by Internet media type text/html type analysis, and give the news of interest to the user based on collaborative filtering technology. The main features of the system include data acquisition, data analysis and storage, information recommendation, information presentation.Collaborative filtering recommendation system is one of the earliest and the most successful technology and its basic idea is:find the correlation of the content or information according to user preference of content or information, or to find user correlation.Recommended research has been very mature, but really want to apply it in the actual scene, also need data collection, and data initialize, no matter how good the recommended model and algorithm can not do anything whithout these two work. The research of recommendation algorithm is based on open source organizations structured data, which is often very simple and clear.However, in practical applications, the state of the data will not be like the experimental data as simple, especially when face the Internet media type text/html type, because this type of data is semi-structured, and the problem is particularly prominent.This paper through the design and implementation of the recommended platform prove the effectiveness of data acquisition, the data initialization methods and the recommended mode to be used to specific media types.