gclen/abcs-of-data-science

abcs-of-data-science's Issues

O is for Outlier Detection | ABCs of data science

O is for Outlier Detection | ABCs of data science

Outlier (or anomaly) detection is the technical term for “finding weird stuff”. It’s used in a wide variety of applications including malware detection and looking for credit card fraud. For example, if you live in Ottawa but your credit card was used to buy a gaming console in Boise, Idaho (without any other purchases) that would be anomalous. Outlier detection is related to clustering. In clustering we are trying to find the groups of related data. In outlier detection we are trying to find the points that don’t belong to any groups. There are three different categories of outliers

https://abcsofdatascience.ca/blog/o-is-for-outlier-detection

Y is for You Should Talk to Your Clients | ABCs of data science

Y is for You Should Talk to Your Clients | ABCs of data science

Many of the previous blogs in this series have focused on the more technical aspects of data science like training models and measuring performance. In all real world data science projects, you’ll need to do more than just the technical aspects. Unless you’re doing it to learn (a perfectly valid reason!), we typically aren’t doing data science for the sake of doing data science. We don’t want to spend tons of time and effort on a data science project, just to have it dropped on the floor or ignored. It’s crucial to communicate with everyone involved in your data science projects to increase the chances of your projects being adopted/recognized.

https://abcsofdatascience.ca/blog/y-is-for-you-should-talk-to-your-clients