Based On The Rapid Classification Of The Interest In The Decision Tree Algorithm Is Optimized

The data mining play an important role in scientific research and business application. With the increase of data quantity, the problem of ability to deal with the data of instrument which used for data mining seems outstanding day by day. The data mining and usually have another name called the knowledge of the database to find. For systematic mining the decision of technical application and enterprise of the data, change the data resources of enterprises into the key competitiveness of enterprises, an effective method is to combine mining technology and enterprise's knowledge base technology together organically, form a system which include research of analyse and demand of application promote mutually, the method to drawing the rule and knowledge professionally.It implies the useful knowledge behind data that the data mining is obtained from the data. The data mining used to integrate the historical data, set up the model of mining on this basis, mining out valuable business rule and mode. Then shows them into an intelligible rule, and integrated in enterprise's knowledge base. The last task is apply the knowledge base in the business activity of enterprises. Will produce different kind of knowledge in different data mining task. Through the research on structure and nature of these knowledge, can get the corresponding information in other task of data mining which have same rule. Thus define the norm in the procedure of algorithms of data mining.The decision tree had a lot of algorithms, this paper focus on the optimization of fast classification in the face of n-value attribute of ID3 algorithm which had parameters of user's interest. On the basis of avoiding the weak relevant attribute of n-value covered the worth strong relevant attribute, simplify complexity of the original algorithm and code cost through the mathematics tool, thus raise the speed of operation while using this algorithm, and lower costs in thrift as much as possible, to raise the efficiency.