I recently read Why Data Science Projects Fail: The Harsh Realities of Implementing AI and Analytics, without the Hype. It is a great book that talks about what why many data science projects fail. It is a very informative read and I suggest you check it out. In this post I will give a very quick summary of what I learned for my future self. Read more

ARIMA is a well-known model for time series forecasting. In this post we will go over the building blocks: MA(q) models, AR(p) models and ARMA(p,q) models of such a powerful model. Read more

Attackers maintains fake accounts to use them for various reasons, for example, in a social media website, attackers could use them to post content to push a certain ideas. Another example in online retail where attackers use fake accounts to add fake reviews to improve the visibility of certain product. In this post we will cover random walk based model proposed in Random Walk based Fake Account Detection in Online Social Networks to detect fake accounts. Read more

Buy-it-again is a popular recommendation model in online retail. In this post we will cover a simple model proposed in Buy it again: Modeling repeat purchase recommendations that achieves good results despite its simplicity. Read more

These are my notes for the lecture Networking with Purpose by Dr Peter Kelly I suggest you check it out if you have the time. Read more

The Root Mean Square Error (RMSE) is a common loss function used in machine learning for regression tasks that measures the average distance between the predicted and true values. However, what implicit modeling assumptions are me making when we use such error to evaluate the model fit? Read more

Suppose we we have sorted array and we are interested in finding a particular element. There are a number of ways we can go about it. Read more

Many of you already know that to in computer systems, to represent the negative numbers we use what is known as 2’s complement. In this article I will try to explain where it came from. Read more

The Luhn formula is a simple checksum formula used for validating identification numbers. It is designed to protect against accidental errors in data entry but not malicious attacks. For example, on a website a user may enter a credit card number or ISBN and sometimes these numbers may be mistyped. One can then use the Luhn formula to determine if the entered number is a valid one. Read more

K-means is a clustering algorithm that is relatively old yet it still in use today because of its simplicity and power. Read more