Observation bias and sub-group differences can easily produce statistical paradoxes in any data science application. Ignoring these elements can therefore completely undermine the conclusions of our analysis.

A/B testing is one of the most popular controlled experiments used to optimize web marketing strategies. Statistical hypothesis testing is applied to compute the statistical significance of the collected data.

From knowledge graphs to social networks, graph applications are ubiquitous. Deep learning models can be successfully applied to graphs data structure by generalizing the concept of convolutional neural network.

The best way to improve the performance of a "SELECT" query is to create indexes on the columns of interest. Knowing about Clustered indexes, "FULLTEXT" indexes, "UNIQUE" indexes can make your index even more effective.

Using aggregate functions and window functions with the SQL clauses "GROUP BY" and "PARTITION BY", respectively, allows us to compute values that depend on several rows in the table.

Tracking datasets and outputs of Data Science and Machine Learning projects with a Git-like interface.