Generative AI, particularly in the form of Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), has gained significant attention and utility in algorithm development across various domains. Here’s how generative AI is being used in algorithm development: 1. Data Augmentation: Generative models can create synthetic data that closely resembles real data. This synthetic data can …

The Use Of Generative AI In Algorithm Development Read More »

Converting data from one scale to another is such a common task for data scientists. This blog post is beyond sklearn’s capabilities. I have always lacked one specific scaler, so in this post I am writing it myself – ScoreScaler. Consolidating data despite different scales It is very common to compare scores or product ratings. …

Transforming from one scale to another Read More »

On September 24, the Germans will elect a new federal parliament. In this lesson, I text the campaign manifestos of the main parties, extract the hidden semantic space and visualize it to see who is closer to whom in German politics. How close are Germany’s election manifestos in meaning? If you want to know what …

How close are German political parties to each other? Using R to derive the latent semantic network of German election manifestos Read More »

Star Trek is a very rare positive vision of the future. Which film attracts the audience the most with this encouraging message? I studied python to try every movie ratings I could find to answer this question. If you follow this guide, so can you. Which Star Trek movie is the best? The short answer …

Using Python, the IMDb API, and web-scraping rotten tomatoes to find the best Star Trek movie Read More »

When evaluating binary classification algorithms, it is recommended to have a baseline for performance metrics. In this blog post, I am calculating the classification efficiency of really stupid classifiers. These models do not use any feature information. If your own classification model works the same way, then a problem arises. Dumb classifiers When assessing how …

The surprisingly good performance of dumb classification algorithms Read More »

The doubling time of Covid-19 cases has become one of the key indicators of the corona pandemic. Political decision makers use this number to decide when isolation measures should be relaxed. In this post, I show that different assumptions about the virus outbreak lead to different doubling time estimates. Which number should you trust? Covid-19 …

The tricky question of how long it takes for Corona cases to double Read More »

2018 was a wild summer for the bike rental market in Berlin. In the beginning, many new bicycle systems were introduced to the market. By now, two have already left. In the meantime, I’ve counted every rental bike I saw. Which bikes are rented the most? Berlin bike rental market in 2018 On May 7, …

An analysis of the rental bike market in Berlin Read More »

A little over a year ago, I decided to pursue a career in data science. Today I work as an Educational Data Scientist at StackFuel, a small startup in Berlin. How did I do it? Timeline of Evolution of Data Science: A Year in the Life In September 2016, I left the Netherlands, where I …

Starting off in data science Read More »

Massive Open Online Courses (MOOCs) have not revolutionized education. Why? They suffer from terrible completion rates. Most students start a MOOC before completing it. In this blog post, I’ll take a look at what my own company’s eLearning completion rates would be if we offered standard MOOCs. How many people go through MOOCs? When I …

Predicting typical completion rates of online courses Read More »