基于数据挖掘和神经网络的赛马决策与分析

运用了数据挖掘技术和神经网络技术，对赛马数据信息进行自动获取和智能分析。首先，分析了目前赛马的有关数据，按照现存的数据信息创建相关的数据库；在数据的搜集方面，建立了一个提取器，在一些网页中读取有关的数据，大大提高了系统自动获取数据的能力；在分析与预测方面，结合了神经网络逻辑建立一个数学线性回归方程，其功能是通过已有的大量的数据资料预测出将要进行的赛事结果，即参加该场赛事的所有马匹的成绩排名。系统主要包括管理模块、数据查询模块、预测模块3大模块，其中预测模块是核心，它是利用对赛马结果有一定影响的13个因素进行数据分析，得出一个预测时间，再与由优先级数据筛选得出的时间值通过两个调节参数做出修正，得出最终的预测结果。

Horse racing decision and analysis based on data-mining and neutral network

In the process of establishing the system the KDD technology and neural networking technology have been integrared to realize the automation of import data from different homepages and the intelligence of analysis of data management. First, this text analyzes the situation of the data management in horse racing field also and establishes an extractor to collect the data from several websites according the data source to construct the relative database. The capability of automation of import data has been increased greatly. In the forecast and analysis model, a linear regression equation has been built by using the neural networking logic. The main function is to find out the prediction of horse racing through analyzing a large data. The system includes the management model, query model, forecast model. And the forecast model is principal, which analyzes the data with 13 factors that influence the result to obtain the predicted racing time. And at last, the final predicted time will be carried out by comparing and adjusting with the parameter of priority data search.