The Design And Implementation Of A Wireless Modeling System Based On The WEKA Data Mining

Under the wave of information technology development, Individual enterprises haveset up their own computer services platform, Also begin to pay attention to theimportanceof information, and focusing on information collection. The size of theinformation is in the growth of the geometry type, The amount of data is too large so thatthe analysis of these data becomes difficult, And the information hidden in a high value.How to make full use of these massive data analysis in valuable data, thereby improvingthe efficiency of the company, to make a significant contribution to the development of thecompany.Data mining technology has provided technical support for the research in this area.Based on the research of the data mining technology and detailed analysis of the projectrequirements, Make The data mining technology combine with the company’s businessneeds, and give up a first data warehouse, and preprocessing data warehouse, and finallycreate OLAP model to analyze business intelligence data mining process.Modeling system for people who do not have professional knowledge of data miningcan also make a simple model for the purpose of analysis, use the Web Interactive Datamining to Increase the user experience, at the same time use Weka open source miningalgorithm to reduce development costs. With the combinationofData mining and Websystems realize the construction and operation of the modeling system In MVCdevelopment mode,Applying data mining techniques to the modeling system, the modelresults analysis.In the data mining,in other words,in the model analysis phase. First, to build theirownmodel, add the appropriate indicators. These indicators data extracted from the stock tosee network data stored in the modeling database data only determined, and finallythrough the automatic and manual removal of discrete points, remove data outliers,making the analysis more accurate results. By clustering and classification algorithms,analysis and found that the presence of anomalies in the data, and timely analysis of thedeviation from the data. Using linear regression algorithm to analyze the relationshipbetween the two indicators to identify their intrinsically linked.Finally,be released in the system, the more people involved in the analysis of this model, found that the probabilityof the model problem and improve the quality of the analysis of the model.Finally, according to their own practical experience in the mining project, summarizethe work in the project and asked the improved portion of the Web-mining algorithmpoints.