Web Data Mining Reserrch And Application Based On Flower Maket

With the rapid popularization of database technology and computer network, using data mining technology, mining under the large data of deep relationship has made marketing more accurate and more effective. It has become a hot topic. The flowers and plants industry of our country grows and expands with Reform and Opening up, but now, the flower e-commerce applications is still at the early stage of exploration. Therefore, Data mining technology applies to the field of flowers and plants e-commerce has more application value.This paper, according to the characteristics of the flowers and plants e-commerce industry, combined the crawler technology, database technology, descriptive statistical analysis, text mining and text visualization technology, with flowers and plants e-commerce. The author uses Chinese online flower websites as the research objects, through the flowers products analysis of related factors to understand the flower market, analysis of the website the classification of products, price, product distribution, language of flowers characteristics, etc., to mine the website’s product classification layout, product price ratio distribution, language of flowers features of different productssuch as implicit data information, provide data support for flower enterprises in order to make precise marketing and decisionmaking.This paper introduces the background and purpose of the related research, summarizes current status and technology of the related research, respectively for data capture and storage, data processing, data analysis and text mining to study the three research stages, and gives the table structure of the MySQL database after structured processing and some core algorithm codeofRlanguage program.First, the paper through the multi-analysis of parsing the target flower website, using XPath positioning extraction related product information, stored in the MySQL database.Then, data processing of semi structured data and text data. Among them, semi structured data processing includes removal of duplication, noise removal, data type conversion, missing value processing and other operations. It also includes text data preprocessing such as Chinese word segmentation, text conversion and generate Term-Document matrix etc.Finally, the data analysis and text mining. Through analysis of the frequency analysis and cross analysis of structured data, we can see basic distribution features of the data, analyze the relationship between variables, use the chi-square to test hypothesis. After text data preprocessing, implementation of association analysis and cluster analysis, the author gets a summary which can be shown in text visualization technology Word Cloud, network of terms diagram and cluster tree.