基于文本分类的农业种植信息集成推荐方法研究

目前网络上存在着海量的农业信息,但是对于广大农民来说信息得不到有效的利用,迫切需要对信息进行集成推荐.针对网络上的农业种植方面的文本信息进行了深入研究,该系统首先利用爬虫技术自动地爬取海量农业种植信息,经清洗整理后构建数据集语料库.其次利用机器学习中KNN方法找到每个样本的k近邻对文章进行聚类,通过TF-IDF方法提取出关键词并构造词频矩阵,然后从文本中构建特征向量,进而对相似文档进行分类,最后将加权值经排序后的结果推荐给用户.该系统实现了对农业文本进行准确的自动分类以及自动提取出文章摘要,并对相似文章进行推荐展示的效果.

Research on the method of agricultural planting Information integration recommendation based on text classification

At present, there is a lot of agricultural information on the network, but for the majority of farmers the information is not effectively used so the urgent need for us is integrate the information and recommend them to the farmers. This paper makes a study on the information of agricultural planting on the network. The system uses the python to automatically crawl large of agricultural planting information, and then builds the data corpus after cleaning. Secondly, the KNN method is used to find the k-nearest neighbor pairs among the news. The key words are extracted by TF-IDF method and the word frequency matrix is constructed and the feature vectors are constructed from the text, and then the similar documents are classified. Take the recommended results to the user. The system realizes the automatic classification of the agricultural news and automatically extracts the abstract of the article, and the similar articles to recommend the effect of the show.