Building Decision Trees Research Articles

Review of open libraries for pharmacoeconomic analysis in R environment

Background. To reveal the full complexity of the relationship between medical intervention and disease outcome, new methods of analysis and modeling are actively being developed, and tools are becoming more complex, for the use of which it is important to understand their limitations and advantages.Objective: conducting a critical review of the main open-source packages in R environment for conducting pharmacoeconomic analysis.Material and methods. The selection of libraries used for pharmacoeconomic analysis in the R environment was carried out based on the keywords “health economic”, “DALY”, “QALY” in the CRAN repository. Only libraries that were valid on the date of the review were included in the study. The selected 10 R software libraries for pharmacoeconomic analysis were reviewed from the standpoint of the number of tools they support, the format of the data used, the possibilities of visualizing results and generating reports, the presence of vignettes and the possibilities of parallelizing calculations.Results. The selected libraries can be divided into three classes: packages for calculating various quality of life indices, libraries for calculating indicators and indices of economic effectiveness of medical interventions (DALY, QALY, ICER), libraries for performing sensitivity analysis of the effect of medical interventions based on decision tree algorithms and Markov models. The libraries “heemod”, “hesim”, “rdesign” allow building simple Markov and semi-Markov models, but preference should be given to “heemod” due to the presence of vignettes. To conduct an analysis using cohort Markov models, partitioned survival models, it is recommended to use the “hesim” library, if there are gaps in the results, it is recommended to use “missingHE”. The “rdesigin” library allows building decision trees indicating the risk of developing certain conditions and the cost of therapy. The “survHE” library for survival analysis, used specifically in health economics, allows you to carry out probabilistic sensitivity analysis based on survival models. To calculate the survival models themselves to identify predictors of a patient's transition from one health state to another, you will need to additionally install the “flexsurv” library. To visualize the results of pharmacoeconomic modeling, you should additionally install the “diagram” and “ggplot2” libraries. Conclusion. The conducted critical review of open source libraries in R environment can serve as a navigator for choosing a tool for performing pharmacoeconomic analysis.

Belief rule learning and reasoning for classification based on fuzzy belief decision tree

The belief rules which extend the classical fuzzy IF-THEN rules with belief consequent parts have been widely used for classifier design due to their capabilities of building linguistic models interpretable to users and addressing various types of uncertainty. However, in the rule learning process, a high number of features generally results in a belief rule base with large size, which degrades both the classification accuracy and the model interpretability. Motivated by this challenge, the decision tree building technique which implements feature selection and model construction jointly is introduced in this paper to learn a compact and accurate belief rule base. To this end, a new fuzzy belief decision tree (FBDT) with fuzzy feature partitions and belief leaf nodes is designed: a fuzzy information gain ratio is first defined as the feature selection criterion for node fuzzy splitting and then the belief distributions are introduced to the leaf nodes to characterize the class uncertainty. Based on the initial rules extracted from the constructed FBDT, a joint optimization objective considering both classification accuracy and model interpretability is then designed to further reduce the rule redundancy. Experimental results based on real datasets show that the proposed FBDT-based classification method has much smaller rule base and better interpretability than other rule-based methods on the premise of competitive accuracy.

TEACHING AND LEARNING TO CONSTRUCT DATA-BASED DECISION TREES USING DATA CARDS AS THE FIRST INTRODUCTION TO MACHINE LEARNING IN MIDDLE SCHOOL

This study investigates how 11- to 12-year-old students construct data-based decision trees using data cards for classification purposes. We examine the students' heuristics and reasoning during this process. The research is based on an eight-week teaching unit during which students labeled data, built decision trees, and assessed them using test data. They learned to manually construct decision trees to classify food items as recommendable or not. They utilized data cards with a heuristic that is a simplified form of a machine learning algorithm. We report on evidence that this topic is teachable to middle school students, along with insights for refining our teaching approach and broader implications for teaching machine learning at the school level.

Enhancing Groundwater Quality Evaluation Using Associative Rule Mining Technique with Random Forest Split Gini Indexing Algorithm for Nitrate Concentration Analysis

Human actions and changing weather patterns are contributing to the growing demand for groundwater resources. Nevertheless, evaluating the quality of groundwater is crucial. Nitrate is a significant water contaminant that can lead to blue-baby syndrome or methemoglobinemia. Therefore, it is necessary to assess the level of nitrate in groundwater. Current methods involve evaluating the quality of groundwater and integrating it into the models. The inappropriate datasets, lack of performance, and other constraints are limitations of current methods. Ground water dataset is used and pre-processed the data’s. Selected data’s are feature extracted and associated with the rule ranking. In the suggested model, the use of associative rule mining technique has been implemented to address these challenges and assess nitrate levels in groundwater. The method of rule ranking is carried out using association rule mining technique to divide the datasets. The split gini indexing algorithm is introduced in the proposed model for data classification. The Split Gini Indexing algorithm is a decision tree induction algorithm that is used to build decision trees for classification tasks. It is based on the Gini impurity measure, which measures the heterogeneity of a dataset. The quality of groundwater has been classified using Naïve Bayes, SVM, and KNN algorithms. The proposed approach's efficiency is evaluated by calculating performance metrics such as precision, accuracy, F1-score, and recall values. The suggested method in the current research attains an improved accuracy of 0.99, demonstrating enhanced performance.

Remote sensing monitoring of changes in forest cover in the Volyn region: a cross section for the first two decades of the 21st century

The aim of the article. This article highlights the significance of forest cover as an important indicator of the state of the environment. It discusses the findings of the Food and Agriculture Organization of the United Nations (FAO) Forest Resources Assessment (FRA) 2020 report, which states that the world's forest area has decreased by 178 million hectares since 1990. The case study of Volyn region shows how cloud processing and vegetation classification can help quantify forest dynamics from 2000 to 2020, allowing local authorities and decision makers to monitor and analyze trends in near real time. Overall, this work provides insights into the importance of monitoring forest dynamics and the potential for remote sensing technology to facilitate this process. Data & Methods. Remote sensing is an effective tool for monitoring forest ecology and management, and Google Earth Engine (GEE) is an online platform that combines data from various agencies to analyze environmental data. The article presents a case study of the Volyn region and how cloud processing and vegetation classification were used to assess forest dynamics from 2000 to 2020. The study used data from Landsat 7 Collection 1 Tier 1 composites and the CART algorithm for binary decision tree building. The study was based on information provided by the Main Department of Statistics in the Volyn region on the area of forests and areas where logging was carried out during the specified period. Research results. It is interesting to note that despite the decrease in logging activities, there is an increase in forest cover loss within forest ranges. This could be due to various reasons, such as illegal logging or natural disturbances like fires or disease outbreaks. The use of machine learning methods like CART classification can help to identify and monitor these changes, which can then be used to inform policy decisions and management practices to reduce forest cover loss. In general, in the Volyn region, there is a gradual decrease in the areas where various kinds of logging are carried out from 524 km2 in 2003 to 239 km2 in 2020. In contrast, forest cover loss within forest ranges increased rapidly from 37.85 km2 in 2015 to 84.01 km2 in 2017 and beyond from 5.53 km2 to 10.80 km2 in 2015 and 2017 respectively. In this study, the accuracy assessment was performed using 30% of the control points obtained initially, based on data on the reliability of the land cover. The manufacturer's accuracy and user accuracy were calculated to evaluate error omissions and possibilities of a pixel being categorized in a certain category. The spatial resolution of Landsat 7 data used in this study was 30 m, with a minimum calculation area of 0.337 hectares. The overall accuracy and the coefficient κ are the most representative measures of accuracy, with an average accuracy of classification of OAav=98.82% and κav=0.9764.

WEB MINING TO DETECT ONLINE SPREAD OF TERRORISM

To fight online terrorism, integrating web mining techniques with data mining algorithms is crucial. Various algorithms like Logistic Regression, K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Naive Bayes (NB), Decision Trees (DT), Random Forests (RF), and Gradient Boosting (GB) are studied for detecting terrorist activities online, extracting data, identifying patterns, and relevant information. Logistic Regression provides a probabilistic framework, KNN uses similarity metrics, SVM constructs hyper-planes, NB assumes feature independence, DT builds decision trees, RF applies ensemble learning, and GB boosts weak learners' performance. These algorithms aid in proactive monitoring and prevention of online terrorism through efficient analysis of structured and unstructured web data. By combining web mining and data mining strengths, this study emphasizes a comprehensive approach to combatting online terrorism dissemination, helping security agencies anticipate evolving threats and prevent terrorist propagation effectively. Key Words: Terrorism, Naive-bayes, random forest, web mining, Gradient boosting

Determining the Rheological Parameters of Polymers Using Machine Learning Techniques

Introduction. The paper investigates the methodology for determining the rheological parameters of materials based on the nonlinear Maxwell-Gurevich rheological model using the stress relaxation curves. The review of the main directions of the metaheuristic approaches (local search, evolutionary algorithms) to solving the combinatorial optimization problems is presented. The metaheuristic algorithms for solving some important combinatorial optimization problems with the special emphasis on building decision trees are described. The comparative analysis of the algorithms for solving the regression problem in CatBoost Regressor is carried out. The aim of the work is to determine the rheological properties of polymers using machine learning techniques.Materials and Methods. The objects of the study are the generated data sets obtained on the basis of the theoretical stress relaxation curves. The source data tables for model training across all samples are presented, and the statistical analysis of the source data sets characteristics is carried out. The total number of numerical experiments across all samples amounted to 346020 variations. To develop the models, the CatBoost artificial intelligence techniques were used; the regularization techniques (Weight Decay, Decoupled Weight Decay Regulation, Augmentation) were used to increase the model accuracy; and Z–Score technique was used for data normalization.Results. As a result of the research, the intelligent models for determining the rheological parameters of polymers (initial relaxation viscosity, velocity modulus) have been developed based on the generated data sets on the example of the epoxy binder EDT-10. Based on the testing results of the models with the best parameters, the quality assessments were carried out: for the parameter 𝜂∗0 the range of values MAPE 0.46 — 2.72, MSE 0.15 — 1.09, RMSE 0.19 — 0.44, MAPE 0.46 — 1.27; for the parameter 𝑚∗ — MAPE 0.07 — 0.32, MSE 0.01 — 0.13, RMSE 0.10 — 0.41, MAPE 0.58 — 2.72. The resulting metric values are permissible. The training graphs demonstrate the stability of the process.Discussion and Conclusion. The developed intelligent models are scalable and cross-platform, have practical applied significance that ensures their implementation in a wide range of the scientific and engineering apps.

Exploring school factors of academic achievement in Macau: The application of educational data mining in Progress in International Reading Literacy Study (PIRLS) 2016 assessment

International large-scale assessments, provide structured and static data, and due to their extensive databases, they can be considered as a valuable resource for Big Data in Education. In this paper, we propose an educational data mining approach to detect and analyze factors linked to academic performance in Macau schools using data from the Progress in International Reading Literacy Study (PIRLS) 2016. We conducted a secondary data analysis based on a set of socioeconomic, process, and outcome variables from PIRLS and other sources, and built decision trees to obtain a predictive model of school performance. By doing so, we were able to identify the school and student-level variables that are most significant in predicting student performance in Macau. These findings will be useful for informing educational policy decisions and shedding light on the causes of poor performance in Macau schools. Overall, our study highlights the potential of educational data mining approaches in analyzing large-scale assessment data and generating insights for educational research and practice.

Classification Performance Analysis of Decision Tree-Based Algorithms with Noisy Class Variable

Class noise is a common issue that affects the performance of classification techniques on real-world data sets. Class noise appears when a class variable in data sets has incorrect class labels. In the case of noisy data, the robustness of classification techniques against noise could be more important than the performance results on noise-free data sets. The decision tree method is one of the most popular techniques for classification tasks. The C4.5, CART, and random forest (RF) algorithms are considered to be three of the most used algorithms in decision trees. The aim of this paper is to reach conclusions on which decision tree algorithm is better to use for building decision trees in terms of its performance and robustness against class noise. In order to achieve this aim, we study and compare the performance of the models when applied to class variables with noise. The results obtained indicate that the RF algorithm is more robust to data sets with noisy class variable than other algorithms.

ML assisted enhanced recommendation system for reliable volunteering environment

The coronavirus pandemic has negatively impacted everyone around the globe. The need to follow social distancing has increased the demand for technology that makes remote volunteering accessible. Alerting, aiding, and educating individuals at risk in a privacy-preserving manner are the need of the hour. Several NGOs have taken up this responsibility but are lacking the resources to reach willing volunteers and donations. This project attempts to make a novel on-demand service application using ID3, CART, and C4.5 decision tree classification algorithms and compare their accuracies. Algorithm C4.5 here we use as a Decision Tree classifier that can be make decisions based on univariate or multivariate predictor. C4.5 Algorithm builds decision trees based on the conception of information gain, with the decision of each classification associated with the target classification. The best way to determine uncertainty is to use entropy that reflects the basis of decision. We aim to classify volunteer data based on predicted volunteer reliability. C4.5 algorithm shows 11.63 higher accuracy when compared to other algorithm and the value of 0.36 Gini index for the C4.5 indicates that there is an adequate equality. It is conferred that the C4.5 algorithm will best solve this problem based on the required dataset that we have created.

Determining the type of explosive item using a decision tree

The present circumstances in Ukraine delineate specific areas of study. The imperative to identify an explosive device is increasingly paramount for adults and particularly for children. Failing to recognize this danger poses a threat to health and life. The paper presents a decision tree method for classifying the type of explosive objects. This allows you to draw conclusions about the type of explosive object hidden depending on the type of soil. This problem is relevant, as it allows to reasonably explain the type of danger depending on various external factors. The paper offers a general representation of the construction of a decision tree for determining the type of explosive objects. In addition, the work provides a visualization of the division into nodes for a visual representation of the construction of a decision tree, taking into account the division into subtrees and nodes. The nodes are the main attributes: voltage, sensor height above ground level, soil type. The decision tree construction algorithm and the structural block diagram of the ID3 algorithm are given. A study was conducted based on who was chosen to measure the entropy decision tree split chaos. The main characteristics of the ID3 algorithm were listed, taking into account the subject area in which it is used. The paper presents one of the built decision trees, which was built as part of research to determine the type of explosive object depending on the possible type of soil, for example, it is dry and sandy soil. Moreover, the paper provides a comparative analysis of the decision tree model against existing methodologies, shedding light on its efficacy, accuracy, and potential advancements in enhancing threat detection capabilities. This comprehensive analysis not only showcases the model's strengths but also offers insights into potential areas for refinement and improvement

Understanding contributing factors to exoskeleton use-intention in construction: a decision tree approach using results from an online survey

Work-related musculoskeletal disorders (WMSDs) are a major health concern in the construction industry. Occupational exoskeletons (EXOs) are a promising ergonomic intervention to help reduce WMSD risk. Their adoption, however, has been low in construction. To understand the contributing factors to EXO use-intention and assist in future decision-making, we built decision trees to predict responses to each of three EXO use-intention questions (Try, Voluntary Use, and Behavioural Intention), using online survey responses. Variable selection and hyperparameter tuning were used respectively to reduce the number of potential predictors and improve prediction performance. The importance of variables in each final tree was calculated to understand which variables had a greater influence. The final trees had moderate prediction performance. The root node of each tree included EXOs becoming standard equipment, fatigue reduction, or performance increase. Important variables were found to be quite specific to different decision trees. Practical implications of the findings are discussed. Practitioner summary: This study used decision trees to identify key factors influencing the use-intention of occupational exoskeletons (EXOs) in construction, using online survey data. Key factors identified included EXOs becoming standard equipment, fatigue reduction, and performance improvement. Final trees provide intuitive visual representations of the decision-making process for workers to use EXOs.

Using entrepreneurial competencies and action to profile entrepreneurs: a CHAID analysis approach

PurposeEntrepreneurial trait and behaviour approaches are used to identify differing entrepreneurial profiles. Specifically, this study aims to determine which entrepreneurial competencies (ECs) can predict entrepreneurial action (EA) for distinct profiles, such as male versus female, start-up versus established and for entrepreneurs within different age groups and educational levels.Design/methodology/approachThe research was conducted using a survey method on a large sample of 1,150 South African entrepreneurs. Chi-squared automatic interaction detection (CHAID) algorithms were used to build decision trees to illustrate distinct entrepreneurial profiles.FindingsEach profile has a different set of ECs that predict EA, with a growth mindset being the most significant predictor of action. Therefore, this study confirms that a “one-size-fits-all” approach cannot be applied when profiling entrepreneurs.Research limitations/implicationsFrom a pedagogical standpoint, different combinations of these ECs for each profile provide priority information for identification of appropriate candidates (e.g. the highest potential for success) and training initiatives, effective pedagogies and programme design (e.g. which individual ECs should be trained and how should they be trained).Originality/valuePrevious work has mostly focused on demographic variables and included a single sample to profile entrepreneurs. This study maintains much wider applicability in terms of examining profiles in a systematic way. The large sample size supports quantitative analysis of the comparisons between different entrepreneurial profiles using unconventional analyses. Furthermore, as far as can be determined, this represents the first CHAID conducted in a developing country context, especially South Africa, focusing on individual ECs predicting EA.

Machine learning based ischemia-specific stenosis prediction: A Chinese multicenter coronary CT angiography study

ObjectivesTo evaluate the performance of coronary computed tomography angiography (CCTA) derived characteristics including CT derived fractional flow reserve (CT-FFR) with FFR as a reference standard in identifying the lesion-specific ischemia by machine learning (ML) algorithms. MethodsThe retrospective analysis enrolled 596 vessels in 462 patients (mean age, 61 years ± 11 [SD]; 71.4 % men) with suspected coronary artery disease who underwent CCTA and invasive FFR. The data were divided into training cohort, internal validation cohort, external validation cohorts 1 and 2 according to participating centers. All CCTA-derived parameters, which contained 10 qualitative and 33 quantitative plaque parameters, were collected to establish ML model. The Boruta and unsupervised clustering algorithm were implemented to select important and non-redundant parameters. Finally, the eight features with the highest mean importance were included for further ML model establishment and decision tree building. Five models were built to predict lesion-specific ischemia: stenosis degree from CCTA, CT-FFR, ΔCT-FFR, ML model and nested model. ResultsLow-attenuation plaque, bend and lesion length were the main predictors of ischemia-specific lesions. Of 5 models, the ML model showed favorable discrimination for ischemia-specific lesions in the training and three validation sets (area under the curve [95 % confidence interval], 0.93 [0.90–0.96], 0.86 [0.79–0.94], 0.88 [0.83–0.94], and 0.90 [0.84–0.96], respectively). The nested model which combined the ML model and CT-FFR showed better diagnostic efficacy (AUC [95 %CI], 0.96 [0.94–0.99], 0.92 [0.86–0.99], 0.92 [0.86–0.99] and 0.94 [0.91–0.98], respectively; all P < 0.05), and net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were significantly higher than CT-FFR alone. ConclusionsComprehensive CCTA-derived multiparameter model could better predict the ischemia-specific lesions by ML algorithms compared to stenosis degree from CTA, CT-FFR and ΔCT-FFR. Decision tree can be used to predict myocardial ischemia effectively.

Improving McDiarmid Tree Performance for Predicting Heart Disease from Data Streams with Missing and Meaningless Values

Building machine learning algorithms for real-time clinical decision support systems has become a current research hotspot. The success of these algorithms depends on their ability to handle data stream characteristics. For example, we cite as characteristics the large amounts of data, the high speed and rate of incoming data, and the change in data nature and distribution over time. The Very Fast Decision Tree (VFDT) is a method for incrementally building decision trees. Since its proposition in the literature, it has become one of the most popular tools for data stream classification. This paper aims to optimize a new version of VFDT called McDiarmid Tree (MT) for the early prediction of heart diseases in real-time clinical decision support systems. The proposed method for improving MT performance consists of two main mechanisms: detecting the presence of missing and meaningless values in data attributes and handling the impact of this presence. The proposed MT has been compared with MT and VFDT. Simulation results show that the proposed MT attains significantly higher prediction accuracy with less time and model cost (RAM-Hours) than the other two algorithms.

Monitoring Sea Currents with Midrange Acoustic Backscattering

This paper is devoted to an acoustical method of measuring mesoscale sea and ocean currents. Due to the fact that such currents exhibit variability, long-term studies are of great interest. The aim of this study is to prepare a physical foundation to organize current measurements in an automated way using stationary mounted underwater echosounding systems. An acoustic system operating at a frequency of 1–3 kHz (lower than commercial frequencies) that is capable of sensing echo signals from natural inhomogeneities located at distances of 1 to 10 km was tested. The test was conducted during a two-week marine experiment on the northern shelf of the Black Sea. The acoustic system was mounted on a platform together with a weather station and other tools that provided reference values for further comparison. Scattering from moving particles, as well as from wind waves, provides a general opportunity for sensing of currents at remote points. Since most scatterers exist at a depth of at least 2 m or on the surface, the proposed sensing method is going specialized for currents in upper layers. However, analysis of Doppler spectra of the actual returning (reverberation) signal showed that this kind of scattering was mixed with bottom reverberation (which contains no additional frequency shift), and other signal distortions were present. Thus, we proposed a new method of signal processing that is aware of the regional environment. The described method is based on machine learning, namely on gradient boosting to build decision trees, which compute water current properties. Such a computational routine is preceded by an original acoustic signal feature extraction process. Finally, a precision of an order of magnitude was achieved, and a sensing distance of at least 2 km was proven as a result of this study carried out with available instruments.

Local dampening: differential privacy for non-numeric queries via local sensitivity

Differential privacy is the state-of-the-art formal definition for data release under strong privacy guarantees. A variety of mechanisms have been proposed in the literature for releasing the output of numeric queries (e.g., the Laplace mechanism and smooth sensitivity mechanism). Those mechanisms guarantee differential privacy by adding noise to the true query’s output. The amount of noise added is calibrated by the notions of global sensitivity and local sensitivity of the query that measure the impact of the addition or removal of an individual on the query’s output. Mechanisms that use local sensitivity add less noise and, consequently, have a more accurate answer. However, although there has been some work on generic mechanisms for releasing the output of non-numeric queries using global sensitivity (e.g., the Exponential mechanism), the literature lacks generic mechanisms for releasing the output of non-numeric queries using local sensitivity to reduce the noise in the query’s output. In this work, we remedy this shortcoming and present the local dampening mechanism. We adapt the notion of local sensitivity for the non-numeric setting and leverage it to design a generic non-numeric mechanism. We provide theoretical comparisons to the exponential mechanism and show under which conditions the local dampening mechanism is more accurate than the exponential mechanism. We illustrate the effectiveness of the local dampening mechanism by applying it to three diverse problems: (i) percentile selection problem. We report the p-th element in the database; (ii) Influential node analysis. Given an influence metric, we release the top-k most influential nodes while preserving the privacy of the relationship between nodes in the network; (iii) Decision tree induction. We provide a private adaptation to the ID3 algorithm to build decision trees from a given tabular dataset. Experimental evaluation shows that we can reduce the error for percentile selection application up to $$73\%$$ , reduce the use of privacy budget by 2 to 4 orders of magnitude for influential node analysis application, and increase accuracy up to $$12\%$$ for decision tree induction when compared to global sensitivity-based approaches. Finally, to illustrate the scalability of our local dampening mechanism, we empirically evaluate its runtime performance for the influential node analysis problem and show a sub-quadratic behavior.

Evolutionary Learning of Interpretable Decision Trees

In the last decade, reinforcement learning (RL) has been used to solve several tasks with human-level performance. However, there is a growing demand for interpretable RL, i.e., there is the need to understand how a RL agent works and the rationale of its decisions. Not only do we need interpretability to assess the safety of such agents, but also we may need it to gain insights into unknown problems. In this work, we propose a novel optimization approach to interpretable RL that builds decision trees. While techniques that optimize decision trees for RL do exist, they usually employ greedy algorithms or do not exploit the rewards given by the environment. For these reasons, these techniques may either get stuck in local optima or be inefficient. On the contrary, our approach is based on a two-level optimization scheme that combines the advantages of evolutionary algorithms with the benefits of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mathcal {Q}$ </tex-math></inline-formula> -learning. This method allows decomposing the problem into two sub-problems: the problem of finding a meaningful decomposition of the state space, and the problem of associating an action to each subspace. We test the proposed method on three well-known RL benchmarks, as well as on a pandemic control task, on which it results competitive with the state-of-the-art in both performance and interpretability.

Multi-Objective Multi-Instance Learning: A New Approach to Machine Learning for eSports.

The aim of this study is to develop a new approach to be able to correctly predict the outcome of electronic sports (eSports) matches using machine learning methods. Previous research has emphasized player-centric prediction and has used standard (single-instance) classification techniques. However, a team-centric classification is required since team cooperation is essential in completing game missions and achieving final success. To bridge this gap, in this study, we propose a new approach, called Multi-Objective Multi-Instance Learning (MOMIL). It is the first study that applies the multi-instance learning technique to make win predictions in eSports. The proposed approach jointly considers the objectives of the players in a team to capture relationships between players during the classification. In this study, entropy was used as a measure to determine the impurity (uncertainty) of the training dataset when building decision trees for classification. The experiments that were carried out on a publicly available eSports dataset show that the proposed multi-objective multi-instance classification approach outperforms the standard classification approach in terms of accuracy. Unlike the previous studies, we built the models on season-based data. Our approach is up to 95% accurate for win prediction in eSports. Our method achieved higher performance than the state-of-the-art methods tested on the same dataset.

Critical assessment of chemometric models employed for varietal authentication of wine based on UHPLC-HRMS data

The use of metabolic fingerprinting combined with advanced chemometric tools for wine authentication has increased in recent years. Although numerous studies, showing different authentication strategies, have been published, rarely any attention has been paid to the stability of used classification models over a longer time period.Here, we present a reliable and robust metabolic fingerprinting-based multiclass strategy for varietal authentication of wine. Analysis was conducted using ultra-high-performance liquid chromatography coupled to high-resolution tandem mass spectrometry. Two sets of commercial wine samples, one for the creation of classification models (201 wines, five red and five white grape varieties) and one for the verification of their validity over a longer time period (138 wines, three white varieties), were analysed. The generated data from the first sample set were subjected to orthogonal partial least squares discriminant analysis (OPLS-DA). The resulting models were validated and used to build decision trees, which enabled the classification of wine samples according to the grape variety. The individual classification rates of the OPLS-DA models were 90–100%. Overall classification rates of the decision trees were 94 and 96% for red and white wines, respectively. In case of the white wine decision tree, verification of its validity over a longer time period was performed using an additional sample set, analysed four months after the original sample set. From the additional sample set, 87% of samples were correctly classified, thus, the stability of the OPLS-DA classification models over a longer time period was verified. In addition, 25 varietal markers of significant statistical importance, mostly flavonoids, phenolic acids and their derivatives, were tentatively identified.