摘    要:足球作为世界第一大球，自问世就深受广大人民的喜爱。而足球的火热也带动足球彩票的发展。本文将从普通彩民的角度出发，挖掘足彩数据的特点，制定合适的足彩投注策略。

本文首先介绍了足球彩票的发行现状，玩法特点和国内外相关研究，解释了欧洲赔率和回报率的概念和意义，具体地讲解了欧洲赔率和概率的转换关系，以及实际投注时需要注意的事项。其次，本文介绍了2010年1月1日-2018年3月1日的所有比赛的比赛数据集和赔率数据集，讲解了各字段的具体含义，分析了数据集中的特点和规律，并简要介绍了如何通过爬虫的方式获取和实时更新数据集。再次，本文通过分析赔率和概率的关系，利用赔率的特点制定了简单赔率投注法、比较赔率投注法和比较赔率变化率投注法的3种方法的操作准则，利用3种方法进行比较投注，并讲解了参数寻优的方法。然后，本文分析了分类和回归算法和足彩投注的关系，运用分类算法训练的模型预测赛事结果，制定赔率模型投注法。该过程包括数据的预处理、模型的构建、模型的评价和优化，以及模型的运用。最后，本文总结了两类投注法的表现，指出不足之处，以及进一步的改善方法。

最终的实际效果验证了各投注法的可行性和实际操作中的可靠性，且预测结果良好，可供广大彩民参考。

关键词：足球彩票； 欧洲赔率； 分类和预测；数据挖掘

Abstract:As the world's largest ball, football has been loved by the majority of people since its inception. The fiery football also drives the development of football lotteries. This article will also explore the characteristics of Football Lottery data from the perspective of ordinary lottery players, and formulate appropriate football betting strategy.

This article first introduced the current situation of the issue of football lottery tickets, the characteristics of gameplay and related research at home and abroad, explained the concept and significance of European odds and return rates, explained the conversion relationship between European odds and probability, and the need to pay attention to the actual betting. Matters.Secondly, this article describes the game data sets and odds data sets for all competitions from January 1, 2010 to March 1, 2018, explains the specific meaning of each field, analyzes the characteristics and laws of the data sets, and briefly Describes how to get and update datasets in real time via crawlers.Thirdly, this paper analyzes the relationship between odds and probability, and uses the characteristics of odds to formulate the operating principles of three methods: simple odds betting method, comparative odds betting method and comparative odds change rate betting method. Compare bets and explain how to optimize parameters.Then, this paper analyzes the relationship between classification and regression algorithms and Football Lottery betting, uses the model trained by the classification algorithm to predict the results of the tournament, and draws up the odds model betting method. The process includes data preprocessing, model building, model evaluation and optimization, and model application. Finally, this article summarizes the effects of two types of betting methods, points out the inadequacies, and further improvement methods. 

The actual effect also verifies the feasibility of the betting method and the reliability of the actual operation, and the prediction result is good, which can be used for reference by the general public.

Keyword: Football Lottery； European odds； Classification；Data mining

目录

1.引言 7

1.1背景介绍 7