中国的房屋价格一直是一个令人关注的问题，新房的价格会受到宏观因素和微观因素的共同影响。而在二手房市场，它的价格在短期主要受微观因素影响。同时，二手房市场较为混乱，卖房者无法客观地估计待售的房屋价格，买房者无法从市场中找到性价比高的二手房。所以，研究二手房价格的影响因素、确立合理的二手房价格评估模型、寻找性价比高的二手房具有重要意义。因此，本文通过爬虫方法以及查阅相关资料获取数据，再经过数据清洗后得到10197条杭州市二手房售卖数据，采用基于Lasso算法的Hedonic模型研究影响二手房价格的因素，再结合数据挖掘方法与Hedonic模型，帮助买房者找到性价比高的二手房。 通过上述研究发现，影响杭州市二手房价格的因素很多。对于二手房价格有显著的正向影响的因素比较多，有面积、装修程度、产权年限、教育质量、金融配套设施、医疗配套设施、交通便利程度、空气污染情况、区域发展水平、安全程度。而房屋存在时间、楼层位置、噪声污染对于二手房价格有显著的负向影响。在比较了6种不同的数据挖掘模型后，本文发现随机森林和支持向量机在测试集的预测准确率最高并且较为稳定，所以采用这两种模型进行预测，帮助买房者找到了性价比高的二手房，其中支持向量机找到了99套、随机森林找到了281套。本文再结合Hedonic模型以及随机森林和支持向量机，帮助卖房者建立了客观的二手房价格评估模型。

China's housing prices have always been a matter of concern, and the prices of new house will be influenced by both macro factors and micro factors. In the second-hand housing market, the price of second-hand housing is mainly influenced by micro factors in the short-term. At the same time, the second-hand housing market is more confusing and sellers cannot objectively estimate the price of houses for sale. Home buyers cannot find cost-effective second-hand housing in the market. So, it has great significance to study the influencing factors of second-hand housing prices, establish a reasonable evaluation model of second-hand housing prices, and find second-hand house that are cost-effective. Therefore, this article collects second-hand housing data of Hangzhou through reptile method and consults relevant literature, after data cleaning, we obtained 10197 records, and then we adopts Lasso algorithm-based Hedonic model to study the factors that affect the price of second-hand housing, and then combines data mining model and Hedonic model to help buyers find some cost-effective second-hand houses. Through the above research, it is found that there are many microscopic factors affecting the price of second-hand housing in Hangzhou. There are many factors that have a significant positive impact on the prices of second-hand housing, including area, decoration, property rights, educational quality, financial facilities, medical facilities, transportation facilities, air pollution, regional development, and safety. The existing time of housing, floor location, and noise pollution have a significant negative impact on second-hand housing prices. After comparing six different data mining models, this paper finds that the random forest and support vector machine have the highest prediction accuracy and stability, so we use these two models to predict, and help home buyers find cost-effective second-hand house, Among them, support vector machines found 99 sets of second-hand housing and random forests found 281 sets of second-hand housing. Combined with the Hedonic model and random forest and support vector machine, this article helps sellers establish an objective second-hand house price evaluation model.