The Design And Implementation Of Data Presentation Tool Of Weibo Recommendation System Based On Codelgniter

The weibo recommendation service refers to recommending users, weibo, topics, music and so on that user maybe interested to the designate user. As the amount of weibo users and weibo contents is mass, Weibo Recommended system realized automatic recommendation. That if the recommend system reached the expectation and if the result of recommendation is in accordance with the product definition should be tested. However, Weibo Recommended system architecture has many levels. In the bottom of the interface changes recommended tests, often can only be tested on the interface itself, but the impact on the user level interface changes are difficult to visually observe. Testers and product managers is difficult to evaluate the effect of the underlying interface modified directly.The paper is based on the intership in weibo recommemdation group, during which the main task is develop weibo recommendation data presentation tool which shows the underlying influence on the final results of the front-end interfaces to the recommended system, and helps recommended data testers view debugging information and scoring the result data after interface changes. And also, the system helps testers with user -level regression testing, and helps product managers to achieve an accurate grasp of the product.The main work of the paper include:1. Analysing and Modeling the data presentation tool of weibo recommendation system based on UML, and buiding quick prototype to find out the function requirement, and also analysing non-functional requirement.2. Designing the data presentation tool of weibo recommendation system based on the MVC design model. According to sequence from the top to the bottom, the paper design system’s modules’ interfacesfirst and then function of modules, classes’ interfaces and important classes and so on.3. Implements the process:The project development language is PHP. The paper compare different PHP development frameworks, and choose Codelgniter. The system deploys in the LAMP development.4. Testing and optimizing the data presentation tool of weibo recommendation system. And the tool is tested with function test and non-function test. The paper used the curl concurrent request and Memcached technology to improve the performance based on the test results.