Please look through the available Kaggle competitions at: https://www.kaggle.com/competitions and select 1 project.
1. Summary (1 pages)
This should be a summary in your own words of the problem, data, preprocessing techniques you used and any initial observations. When you describe the data, please provide description of atleast 10 features; Description of class label — please try to include tables instead of showing the code output.
2. Benchmarking of Other Solutions (2 pages)
Identify 3 other Kaggle solutions completed by others. The solution should include a score on the Kaggle prediction task. You can find by selecting on the project and then clicking on the link to Kernels. Summarize the features, modeling approach, and performance in a table. Then do
further research to comment on the approach and try to characterize what makes the kernel successful than others. At this stage, you are not expected to build your own model yet.
3. Data description and Initial Processing (3 pages)
This section should include basic characterization of data. You should run and report basic statistics on the data and generate at least 3 visualizations. You can review other kernels to understand some different approaches to the data, but this section you are required to generate all analyses. In the preprocessing, state clearly what has been done to make sure data is ready to build a model – including important visualizations/tables. Please check if these visualizations are helping understand your data better.
