Date of article publishing: February 14, 2021

Article Summary

Companies can use machine learning models to their advantage to predict and attempt to remedy customer churn. In order to do this, data is gathered to learn about a customers gender, join date, probability of contacting customer services, and country of residence. After attaining this data, the information is merged, duplicates are thrown out, the join dates are organized (to establish any seasonal relationships) and customers are put into test and train datasets. The train data sets are customers with a churn flag which indicates loss of that client. Test datasets are for customers without a churn flag. After running the model, outlying and unexpected data points are found and corrected. By using machine learning models in the way described by Ade-Ojo, companies can find correlation between join dates and customer turnover and from there attempt to make an effort to prevent it.

Author Information

John Ade-Ojo is a data scientist working in consumer finance pricing. According to his LinkedIN, John has spent the past 5 years working as a senior analyst for the Royal Bank of Scotland Business as well as Shawbrook Bank. Prior to this, John worked as a risk analyst for HSBC Retail Banking and Wealth Management.

Areas of application

The most clear realm of application for machine learning predictions of customer churn is in a business setting. All companies benefit from being to prdict and mitigate customer turnover. This could also be relevent in a school setting. These models could be used to predict subjects that are most commonly failed and schools can use that knowledge to evaluate the cause of failure and attempt to counter it.

Reflection & Similar articles

## [1] "Personal article rflection and links to similar articles"

My thoughts on the topic

I feel like this article offered great insight for readers of all coding experience levels into the world of machine learning. By doing an example as he described the purpose and method of machine learning models in the business world, Ade-Ojo made it very easy to follow along and visually grasp different concepts. In terms of the information presented, I felt like machine learning offers the ability to play a major role in determining company decisions by predicting the best and worst times to implement new strategies as well as showing where more effort needs to be placed towards preventing customer churn.

Relevant images