Calculate customer experience using machine learning models

Mobile network operators are keen to improve their customers experience in their network, but often face difficulties to measure it for the whole subscriber base. Your task will be to come up with a machine learning solution to predict the customer experience index for all subscribers, based on network KPIs (such as call success rate, data bandwidth, network coverage, …) and customer satisfaction survey results.

The Mobile Network Operator collected the results for 2 million customer satisfaction surveys, where they received an overall score as customer satisfaction (0-10).

They also collected network KPIs for all of their subscribers (2,3 million) for a 2 weeks period, just before the satisfaction survey was taken. They aggregated the most important KPIs and created a profile table.

Your will receive:

  • the results of customer satisfaction survey (a satisfaction index from 0 to 10 for all subscribers who participated in the survey), and
  • the profile table (a list of aggregated KPIs for all subscribers in the Mobile Network Operator’s network).

Your task will be to predict the customer satisfaction survey result for all subscribers in the network.

Here's the data!

CustomerExperience MachineLearning

Criteria

  • Data preparation
  • (3 points)
  • Identify which KPIs are relevant for the Customer Satisfaction score?
  • (2 points)
  • Model accuracy
  • (5 points)

Submission

You will need to provide the predicted customer satisfaction scores (0-10) for all subscribers and also share the source code of your ML modell.

The predicted satisfation scores should be in CSV format as:

<subscriber_id>,<customer_satisfaction_score>