Design And Application Of Data Platform Based On Cluster Analysis

Clustering analysis is a very wide range of data mining technology, clustering analysis or clustering is to group the objects in a certain way, so that the same group (cluster) of the object similarity is high, while trying to make different groups (clusters) between the low object similarity. Clustering is an important task of data mining, and it is one of the most common methods of statistical analysis. In image analysis, intelligent machinery manufacturing and biological information and other fields are used.Modern society, the data is another kind of wealth, banking, telecommunications, retail and other industries are the information on the phone as well as processing. Data mining to transform data into useful information, knowledge, and economic benefits. Each year, the college enrollment, each student in the school's academic performance, is a large number of data, which contains a lot of effective information, the test data for effective clustering analysis, can get effective information, to different classes of learning level can have a horizontal comparison, so as to achieve the teacher's teaching activities.Data platform is a platform for clustering analysis of test data. It is attached to the data mining platform WEKA, a platform of MYSQL database. First, the different storage formats of the data stored in the database has been set up to establish a good database, through the WEKA platform to connect the database, from the database to remove the data to be processed. By cluster analysis, the k-means algorithm, the test data of different classes cluster analysis and the cluster analysis results are analyzed, for different classes to study the level of a horizontal comparison, teachers can according to the comparison result, adjust the teaching methods, according to the different class, different levels of students teaching students in accordance of their aptitude.Data mining technology has not played its due role in the education system, University data storage time, wide range, the majority of information is accurate, but not a good analysis system and can only be used as a backup and access, in fact, the phenomenon is particularly serious in our country.