1. Pilot-Study Proposal Imagine that you work as Machine Learning independent consultant, providing scientific advisory and consulting services to companies seeking to apply data analytics to their business activities. The manager of a large hotel chain contacts you to investigate the feasibility of using machine learning to predict whether a new hotel opened in a given location will be profitable or not (i.e. whether it will make a profit or a loss, regardless of the amount). The manager has access to historical data of successful and unsuccessful hotels opened under the chains brand and she offers to provide you with geographical and socio-economical data about the locations and neighbourhoods. In the first part of your assignment, you are asked to write a detailed proposal for a pilot study to investigate whether machine learning procedures could be used to successfully solve this problem. Your report should discuss several aspects of the problem, including the following main points: the type of predictive task that must be performed (e.g., classification, regression, clustering, rules mining, ...); examples of possibly informative features that you would like to be provided with (what type of information that the company could obtain about the location is likely to be a good predictor?); the learning procedure or procedures (e.g., DTs, k-NN, k-means, linear regression, Apriori, SVMs, ...) you would choose and the reason for your choice; how you would evaluate the performance of your system before deploying it. 1 You can assume that the manager has some knowledge of machine learning and you do not need to explain how the recommended learning method works. Simply discuss your recommendation and back it with sound arguments and/or references. This document should consist of approximately 500{750 words of narrative (i.e. excluding references, pictures, and diagrams). Please report your word count on the title page. The document must be submitted in PDF format with file name CE802_Pilot.pdf.