METHODOLOGY
The CRISP-DM (Cross Industry Standard Process for Data Mining) methodology is proposed for use in this research work. The method is preferred for most data mining projects being done today. The choice in methodology is due to the aspect of the CRISP-DM being independent of any industry domain and technology that can be used to implement the data mining solution. It comprises five steps to follow while doing the work.
In the first step, the researcher has to set goals for the project or work being done; this will help develop the objectives of the work being done. Second step, the researcher has to specify the data and sources they will use in their work. The second step is essential as it defines the type of data that shall be used in the project, whether it meets the requirements of the project, and achieves the project’s intended results. The third step would be data preparation and transformation to make the data ready and usable for the project. The fourth step would be to model the algorithm for my proposed solution and use it on the data I had gathered in the previous three steps. The fifth step will be to evaluate the results of my proposed model based on the accuracy and interpretability of the model I had deduced in step Four. Lastly, the sixth step would be to have a conclusive end to the research based on the evaluation findings of step 5 and comparison with other methods in the same area (Wirth et al., 2000).