Statistical model checking offers an alternative to traditional model checking for large stochastic systems, addressing state space explosion and approximating quantitative properties. This paper proposes machine learning approaches using decision trees to approximate zero-reachability states, offering both computational efficiency and interpretability. Statistical analysis is used as an alternative approach to establish simulation run length bounds to control computation errors. Experimental results across standard Markov models demonstrate that our decision structures maintain high correctness (99% in most cases), reduce runtime, and have minimal memory overhead. Even when some methods show limitations, alternative approaches within our framework yield effective results.
This research presents a novel hybrid portfolio optimization framework that combines the Hierarchical Risk Parity (HRP) algorithm with two Multi-Criteria Decision-Making (MCDM) methods, MEREC and WEDBA, specifically to overcome fundamental shortcomings in the standard HRP model. The central goal is to alleviate the chaining problem and resolve HRP’s difficulty in identifying the optimal number of clusters, issues known to negatively affect portfolio diversification and risk allocation. To achieve this structural improvement, the Elbow method is integrated directly into the HRP process, ensuring a robust cluster structure is defined before any weight allocation occurs. The MEREC method is then utilized to calculate objective criterion weights, while the WEDBA approach is employed to assess the financial performance of individual assets within each cluster generated by HRP. This HRP–MCDM algorithm is tested using daily closing price data for stocks on the BIST 100 Index covering the 2018–2022 period. The performance of portfolios generated across seven distinct linkage methods (Ward, single, complete, average, weighted, centroid, and median) is rigorously benchmarked against the outcomes from the traditional HRP approach. Findings demonstrate that the HRP–MCDM framework significantly boosts both return levels and risk-adjusted metrics, especially when using the single and Ward linkage method, thereby surpassing the standard HRP algorithm in the majority of test cases. By strategically blending machine-learning-based risk clustering with objective, multi-criteria evaluation, this study makes a vital methodological contribution to the portfolio optimization domain, equipping investors with a more stable, transparent, and performance-focused asset allocation instrument.
Journal:Informatica
Volume 36, Issue 3 (2025), pp. 657–676
Abstract
Most classification algorithms involve subjective inputs or hyperparameters to be determined prior to performing the classification. When taking different input or hyperparameter values, each classification algorithm will comprise a collection of classifiers. In this work, we propose a data-driven methodology for assessing similarity in consensus agreement within such a collection of classifiers, and between two classification algorithms, conditional on the dataset of interest. The core of our approach lies in considering the variability introduced by different hyperparameter values for each algorithm when performing such comparisons. We address these problems by evaluating the similarity through consensus agreement and by proposing the application of asymmetric similarity indices based on the Jaccard coefficient. We present the proposed methodology on two publicly available datasets.
Journal:Informatica
Volume 32, Issue 3 (2021), pp. 441–475
Abstract
This paper is devoted to the problem of class imbalance in machine learning, focusing on the intrusion detection of rare classes in computer networks. The problem of class imbalance occurs when one class heavily outnumbers examples from the other classes. In this paper, we are particularly interested in classifiers, as pattern recognition and anomaly detection could be solved as a classification problem. As still a major part of data network traffic of any organization network is benign, and malignant traffic is rare, researchers therefore have to deal with a class imbalance problem. Substantial research has been undertaken in order to identify these methods or data features that allow to accurately identify these attacks. But the usual tactic to deal with the imbalance class problem is to label all malignant traffic as one class and then solve the binary classification problem. In this paper, however, we choose not to group or to drop rare classes but instead investigate what could be done in order to achieve good multi-class classification efficiency. Rare class records were up-sampled using SMOTE method (Chawla et al., 2002) to a preset ratio targets. Experiments with the 3 network traffic datasets, namely CIC-IDS2017, CSE-CIC-IDS2018 (Sharafaldin et al., 2018) and LITNET-2020 (Damasevicius et al., 2020) were performed aiming to achieve reliable recognition of rare malignant classes available in these datasets.
Popular machine learning algorithms were chosen for comparison of their readiness to support rare class detection. Related algorithm hyper parameters were tuned within a wide range of values, different data feature selection methods were used and tests were executed with and without over-sampling to test the multiple class problem classification performance of rare classes.
Machine learning algorithms ranking based on Precision, Balanced Accuracy Score, $\bar{G}$, and prediction error Bias and Variance decomposition, show that decision tree ensembles (Adaboost, Random Forest Trees and Gradient Boosting Classifier) performed best on the network intrusion datasets used in this research.
Journal:Informatica
Volume 12, Issue 3 (2001), pp. 455–468
Abstract
This paper describes a preliminary algorithm performing epilepsy prediction by means of visual perception tests and digital electroencephalograph data analysis. Special machine learning algorithm and signal processing method are used. The algorithm is tested on real data of epileptic and healthy persons that are treated in Kaunas Medical University Clinics, Lithuania. The detailed examination of results shows that computerized visual perception testing and automated data analysis could be used for brain damages diagnosing.