A data miner wants to conduct a classification for a given dataset (D1). The miner wonders the better data mining classification algorithm and configuration parameters for D1, and the miner starts the data mining assistant.
1. The user is asked to provide the type of task. In this case, it is a classification problem.
2. The system automatically extracts the relevant metadata from D1.
3. The user is asked to provide some application restrictions. In this case, the model has to be as accurate as possible.
4. The system generates a recommendation.
5. The user can execute the proposed algorithms with the predefined configuration parameters and validate the results.
6. If the evaluation is not satisfactory enough, the user can execute the algorithms with different parameters.
7. When the user is satisfied with the obtained results, the system learns the new experience with its corresponding solution.