Data science based methods for customer needs mapping

Question 1

Explain supervised learning

Answer 1

Builds a mathematical model of a set of data that containts both the inputs and the desired outputs

Question 2

Explain unsupervised learning

Answer 2

Unsupervised learning algorithm in data science is where the algorithm is given unlabeled data, that contains only inputs and is tasked to find patterns, relationships, overall structure in the data, like clustering or grouping of data points.

Question 3

Criterias for good datasets?

Answer 3

Answers a relevant question
Has a good variability in the key variable we are interested to predict
Has a good variability in the explanatory variables
Has man instances (rows in the table)
Does not have too many missing values

Question 4

What is clustering in unsupervised learning?

Answer 4

Cluster analysis is the task of grouping a set of objects in such a way that objects in the same group (called clusters) are more similar to eaach other than those in other groups

Question 5

How can we predict the probability of purchase? (demand)

Answer 5

1. From discrete choice modeling
   Goal: Predict likelihood of customer purchasing a particular product

Dependent variables: Often binary and represents whether a customer will buy product (1) or not (0)

Independent var: predictions, may include factors as price, advertising, or other that could influence purchasing decision

Method: multinominal logistics regression

2. From historical product data
   Goal: Analyze relationship between multiple independent variables and dependent variables

Dependent var: quantity of product sold

Independent var: factors that may impact sale, like price, packaging, branding etc

Method: Multinominal regression