Web-Weka Meets D3.js in Web Based Medical Data Mining

Clinical Data mining as an interdisciplinary subfield of computer science is becoming increasingly important in recent years. While, so far, there is not a complete web based medical data analysis and mining platform. Weka is one of the most widely used data mining tools with a variety of algorithms integrated. However, as a workbench, it can not be integrated into specific web based business system. Also lots of defects are detected in our or other practitioner´s daily usage such as preprocessing functions can´t satisfy the actual needs. While the most badly shortcoming, we think, is visulization and it can not integrated into web based business system. This paper presents a web based weka in a real application scenario of exploring the relationship between blood routine examination results and fifteen common hematological diseases. As an innovative point, we combine weka with D3, this strategy not only solving the problem of implementing visualization on the web front-end but also making visualization more powerful than weka original. Also, we integrate our weka into pentaho BI.