murilomirandan / imarket Goto Github PK

Our strategy team has decided that we should pursue an exclusive partnership with one of the biggest accessory makers in the market. We’re in talks with two of them: Belkin and Elago.

The sales team engaged a market research firm to conduct a survey of our existing customers. One of the objectives of the survey was to find out which of two brands of accessories our customers prefer. This information will help us decide the manufacturer to partner with. Unfortunately, the answer to the brand preference question was not properly captured for all of the respondents. For that, we investigated if customer responses to some survey questions (e.g. income, age, etc.) enable us to predict the answer to the brand preference question. If we can do this with confidence, I would like you to make those predictions and provide the sales team with a complete view of what brand our customers prefer.

1. Exploratory Data Analysis (EDA) from a survey to iMarket customers.

2. Run and optimize two different decision tree classification methods - C5.0 and RandomForest - and compare which one works better for this data set.