Our Data Mining App is an application with the intended user base being developers moving from Hadoop to the Apache Spark platform. Our software will allow data file uploading, algorithm selection, and attribute selection to extract the exact information the user requires. These selected attributes will be used to generate equivalent Scala code to be executed on the Apache Spark platform. Our software will be relatively transparent to the user. The primary aspect that sets us apart from existing Data Mining tools is our program’s functionality as code generator, producing portable Scala code that can be run on one computer or an entire data center.

Our Data Mining App will include the following functionalities: Provide a Swing-based, standalone, portable Java application interface. Allow user to upload a CSV file, select a mining algorithm from a populated list, and select data attributes. Generate Scala code according to user’s selected attributes with the intent of execution on the Apache Spark platform. If time permitting Data Mining App will provide additional functionality: Support additional mining algorithms (machine-learning, association, linking). Our Data Mining App will NOT include the following functionality: Our application will not provide data mining services, or execution of the generated Scala code.