Skip to main content
Decrease the numbers!

Losing weight efficiently using Data Science

End 2013: I would like to lose some weight. End 2014: Maybe I should start losing some weight. End 2015: I should definitely lose some weight. A few weeks ago: I need to lose weight. At that point and bought a book and read a lot of literature about this topic. I am following a low carbs diet and this resulted into a weight loss of 6 kg in the first 3 weeks! The key points are summarized in this post. My goal is to eventual apply some machine learning algorithms on the data I am currently collecting. If you tried a diet before or if you have collected some related data, please let me know. I will use it for the machine learning post on this topic. In this post I will summarize the key points from the literature and show some visualizations of the results so far.

Note: this text contains subjective statements. These are only based on my experience. I try to give as much pointers to literature as possible, but is should not be considered as the absolute truth.

Where to start?

We (modern Homo Sapiens) exist for 200,000 thousand years on this planet. Humans eat based on emotions. If you look at predators among all animals, you will rarely see a fat one. These animals do not eat by emotions, but they eat to survive. In the beginning of times, we also were predators. So what did we eat since the beginning of time? Potatoes exist for 7,000 – 10,000 years, so that can be considered as young (relative to the 200,000 thousand years we exist). Nuts are on this planet for 780,000 years. That is also one of the key ingredients of our diet in the beginning of times. Also meat and fish were around before the beginning of the modern Homo Sapiens. It turns out that this diet has a low concentration of carbohydrates. Nowadays, food which is bought from supermarkets contains a lot of carbohydrates. Why?


Glucose. You can find it in almost any product in any supermarket. Why? Because it is addictive. The downsides of glucose are the following:

  • If you eat a product containing many glucose, you will get a glucose dip after a few hours and then you want to get some glucose to fill the gap.
  • For having too many glucose, fat is stored more easily in your body.

So we should definitely stay away from products which contain a lot of glucose and carbohydrates.

Low Carbs diet

In the beginning of our times, we had a low carbs diet. Ketogenesis (a.k.a. “fat burning”) will be activated by a lack of carbohydrates. So this could be a good starting point.

Tracking progress

Please do not rely on your weight measuring device. But measure your fat. Your weight measuring device will give a different answer from time to time and measuring your fat is about constant. This gives more reliable insights into your progress. If we would talk in terms of random variables, you weight measuring device has large variance and the variance of measuring fat tends towards zero. I am measuring my fat by having a tape-measure and by measuring the circumference of my belly, one of my upper legs and one of my upper arms. If I lose weight, I can also track which of my body parts was responsible for this.

Results so far


I collected the total amount of carbohydrates, protein and fat of the food I ate per day. I made sure that I had less than 50g of carbohydrates per day. Sadly, I could not attain this goal for all of the days, but I did it for most of the days.


I also made some histograms for the carbs, fat and proteins. It can clearly be seen that there are two groups of carbs: the ones < 50 (normal days) and the ones > 50 (my cheat days).


Weight & Fat

In the figure, there are three weeks (week 0, week 1 and week 2). The transparent line is a linear regression on the datapoints, so it can be used to predict when a certain goal is reached. You can clearly see the weight loss in this picture!



  • A Paleo Diet (the diet like we had in the beginning of times) ensures a low intake of carbohydrates.
  • A low intake of carbohydrates ensures a process called ketogenesis which makes you losing fat.
  • Throw away any weight measuring device and use a tape-measure for tracking progress.

In the next post of this series, I will give a short progress update and I will give some more insights in my diet.


  1. MU, Song-niu, et al. “Rat Model of Type 2 Diabetes Induced by Streptozotocin and High-Carbonhydrate-Fat Diet [J].” Chinese Journal of Comparative Medicine 2 (2008): 005.
Kevin Jacobs

Kevin Jacobs

Kevin Jacobs is a certified Data Scientist and blog writer for Data Blogger. He is passionate about any project that involves large amounts of data and statistical data analysis. Kevin can be reached using Twitter (@kmjjacobs), LinkedIn or via e-mail: