A data-driven system to make smarter bets

Photo by Thomas Serer on Unsplash

Football is the most popular sport worldwide. It is a global game that connects almost every single person on the planet. It has been a part of my life since I was a child. I’ve watched almost every match so far.

Over the last couple of months, I have been working on a project called betting analytics on sports (football) matches. I want to implement a simple betting framework in Python. It has two parts which are data-preprocessing and exploratory data analysis. …


Making more informed, “better” decisions

Photo by Travel Nomades on Unsplash

In part 1, we did a preprocess of the football dataset. In this part, we perform exploratory data analysis. The dataset contains 79 explanatory variables that include a vast array of bet attributes. The dataset gif is below.


Glossary of common Data Science terms

Photo by Franki Chamaki on Unsplash

Data Science field is such a rapidly expanding growing space because data is an asset that empowers businesses to make better decisions. For every field, Data Science has its own jargon which means that only those in a field understand. In this article, I want to highlight the 9 common Data Science jargon or term that we encounter frequently in the real-life.

Data-driven Approach

The first common term is Data-driven which is a significant approach that progresses by data without intuition or domain or personal expertise. If you are working in a consultant firm, you are engaging in a lot of different…


Feature Engineering for anomaly detection models

Definition of prospect: an apparent probability of advancement, success, profit, etc
Definition of prospect: an apparent probability of advancement, success, profit, etc
Definition of prospect: an apparent probability of advancement, success, profit, etc

My aim in this article is to give some intuition about how to create an advanced feature(s) in customer transaction data to predict anomaly or fraud in credit cards.

Fraud and anomaly detection

The machine learning approach to fraud detection has gained popularity in recent years because instead of rule-based methods, self-learning AI models that continually learning and capable of detecting outlier and anomaly behaviors. In the rule-based systems, the algorithms are written by fraud analysts and they are based on a strict threshold.

Supervised vs. Unsupervised Approach

Even though most of the sources said that supervised data common in…

Yavuz Selim Sefunc

Data Scientist at Turk Telekom, Istanbul Turkey. https://www.linkedin.com/in/yssefunc/

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store