• Genetic algorithms

    I have previously been studying a whole bunch of nature inspired algorithms, or in other words optimisation methods inspired by natural phenomena. In this article, I’ll cover the basics of creating a genetic algorithm (GA). The nature of evolutionary algorithms A GA is an evolutionary algorithm. Evolutionary algorithms are inspired by the mechanisms of biological evolution by means of natural selection (as proposed by Darwin). The mechanisms of biological evolution revolves around modification and propagation of genetic material over time. Parents contribute with genetic material to create an offspring, and along the way there is some probability of natural mutation. In layman’s terms Basically, a GA mimics the evolution of some…

  • Simulated annealing

    I’m currently studying a whole bunch of nature inspired algorithms, or in other words optimisation methods inspired by natural phenomena. The simulated annealing (SA) algorithm is an optimisation method that is able to also solve complex optimisation problems. The strength of SA is its ability to overcome the ‘neighbourhood limit’ that restricts simple optimisation algorithms. A classical example of the applicability of SA, also emphasised in the original article Optimization by Simulated Annealing (Kirkpatrick et al. 1983), is its ability to solve the traveling salesman problem (a well known NP problem where we want to optimise a travel person’s path, so (s)he travels the least amount of miles). Optimisation algorithms in general…

  • Firefly algorithm

    Nature inspired algorithms provide interesting ways to optimise complex functions. In one of my courses, Scientific Computing, we were given a function with no analytically known solution and asked to optimise it with the Firefly algorithm. The Firefly algorithm was proposed by Xin-She Yang in the early 2000’s. The metaphor The algorithm is based on the metaphor of a firefly swarm, on the following assumptions: All flies are unisexual and therefore can be attracted to all other flies. The attractiveness of a firefly is proportional to its brightness – brightness decreases with distance, so flies are attracted to each other based on an apparent attractiveness. If a firefly is not…

  • Statistics on wine tasting reviews

    As a part of one of my latest courses ‘Applied Statistics – from data to results’, we were asked to find any kind of large data set and apply at least one hypothesis test on it. My group and I found a data set consisting of 130.000 wine tasting reviews on Kaggle.com and went to it. We worked hard on the data analysis and visualisation as a team and actually ended up learning a lot about applying hypothesis testing to real life data and the programming of it, but probably even more about the importance of considerate data visualisation. We ended up doing a presentation of our findings in front of…

  • Heavy sand

    As a part of the course “From Idea to Result”, we had to conduct a research project within a research field completely new to us. Sand is an important source of industrially valuable components such as magnetite (FeO) or ilmenite (FeTiO). The project I became a part of closely investigated the magnetic properties of a natural sample of heavy sand from Villingebæk beach (North Zealand). In order to do so, we first separated a magnetic fraction from the sample for further analysis (simply by using a strong handheld magnetic and collect magnetic particles). We then employed a variety of experimental methods to analyse the particles’ chemical components, but our analysis concluded…