Sergei Mitrofanov

Decision tree learning algorithms have a long history, but they are still popular due to their efficiency. Tree construction starts at the root and continues to each leaf node, creating a “near-optimal” tree. One of the key steps in creating a decision tree is the selection of a feature to split at each root node, which affects the classification accuracy. This process can be quite labor intensive. The article proposes a new approach to constructing decision trees based on the use of differential evolution for two-level tree optimization. Differential evolution works in two stages: at the first level, a feature is selected for separation, and at the second level, the threshold value for this feature is optimized. The results of the work were tested on several examples of classification problems.

Decision trees belong to the most effective classification methods. The main advantage of decision trees is a simple and user-friendly interpretation of the results obtained. But despite its well-known advantages the method has some disadvantages as well. One of them is that decision tree learning algorithm build an “almost optimal” tree. The paper considers the way to improve the efficiency of decision trees. The paper proposes a modification of decision tree learning algorithms by retraining the part of tree at every node training. The classification problems were solved to compare standard decision tree learning algorithms with the modified ones. Algorithm efficiency refers to the percentage of correctly classified test sample objects. Statistical analysis based on Student’s t-test was carried out to compare the efficiency of the algorithms.

We consider the problem of detecting anomalies in the directional distribution of fibre materials observed in 3D images. We divide the image into a set of scanning windows and classify them into two clusters: homogeneous material and anomaly. Based on a sample of estimated local fibre directions, for each scanning window we compute several classification attributes, namely the coordinate wise means of local fibre directions, the entropy of the directional distribution, and a combination of them. We also propose a new spatial modification of the Stochastic Approximation Expectation-Maximization (SAEM) algorithm. Besides the clustering we also consider testing the significance of anomalies. To this end, we apply a change point technique for random fields and derive the exact inequalities for tail probabilities of a test statistic. The proposed methodology is first validated on simulated images. Finally, it is applied to a 3D image of a fibre reinforced polymer.

Decision tree is a machine learning algorithm that is very effective in classification problems. Decision tree is a topical algorithm because of the good interpretability of the results of their work. However, decision tree has the drawback that standard algorithms don’t allow obtaining the optimal structure of the decision tree. To solve this drawback, it’s proposed to use the genetic programming algorithm. This algorithm is one of the branches of evolutionary algorithms and has proven itself for the design of intelligent information technologies. Genetic programming, in one of their implementations, searches for solutions in tree space, which is well suited for designing decision trees. Ensembles that are based on decision trees have high efficiency. In this paper, random forest and gradient boosting are considered. In ensembles, it’s proposed to combine decision trees that are designed by the genetic programming algorithm. Algorithms was tested on classification problems.