Machine learning models benefit from scaling up the area of the scale where most data points show differences. In this blog post, I present an exponential scaler that does just that. It increases the lower or upper end of the scale to focus the machine learning model on the differences that matter most. Design a …
Exponentially scaling your data in order to zoom in on small differences Read More »
The new Star Trek series, Discovery, is just around the corner. It’s time to look back at the best episodes of the franchise so far. This tutorial will teach you how to use the IMDb API to get responses yourself. What’s the best Star Trek episode? TL; DR: Star Trek The Next Generation: The Best …
Using Python and the IMDb API to find the best Star Trek episode Read More »
In just a month, the largest country in Europe, Germany, will run for elections. In this short guide, I will write the text of the campaign manifestos of the main parties to visualize the state of German politics. What are the topics of German politics? I don’t know a single person who reads election manifestos. …
Using R to derive the German election manifesto word clouds Read More »
I’ve tried millions of beer ratings on the largest beer rating sites to answer all the beer lovers’ questions. What is the best beer? What’s so good about him? Where is the best beer made? Which beer is best? To answer this question, I collected over 128 million ratings of over 8,000 beers from untappd.com …
Using a machine learning algorithm out of the box is problematic when one class in the training set dominates the other. The Synthetic Minority Resampling Technique (SMOTE) solves this problem. In this tutorial, I’ll explain how SMOTE works, and then how the SMOTE function code works. What problem does SMOTE solve? Machine learning algorithms learn …
SMOTE explained for noobs – Synthetic Minority Over-sampling TEchnique line by line Read More »
From November 2017 to July 2019, I worked for a seed-funded startup in Berlin. Most of this time, I tracked my working hours. In this post, I analyze this data. Start time In about a year and a half working in Berlin, I learned more than during the same time at the university. So, I …
Is it worth putting data in a database if all you’re interested in is speed? It depends on what you are doing with the data. This guide teaches you where to expect the speed advantage of SQLite and R. Tune There are many benefits to organizing your data in a database. Few think of speed …
More than 23,000 Germans study in the Netherlands. Many of them do not understand that when they return to Germany, they will be punished. The reason is discrimination based on foreign classes. What can you do about it? How does Germany punish students from abroad? TL; DR: using the so-called “modified Bavarian formula”. This is …
Streets, squares and places are usually named after influential people. As a consequence, place names give an idea of who shaped local cultures. In this post, I use open street map data to visualize the impact of various historical figures in Germany. Requesting data from an open street map To find out where geographic features …
Using geographical heat maps to visualise cultural influence Read More »
Information with a geographic element is best visualized with a map. However, large regions tend to dominate maps regardless of their actual importance. I will show you possible ways to solve this problem and let you create the correct data map for your own purposes without the need for coding. Problem with standard data maps …