Comparative analysis of machine learning methods for solving the problem of predicting failures in gas turbine engines

There is a lack of rigorous methodological development for descriptive epidemiology, where the goal is to describe and identify the most important associations with an outcome given a large set of potential predictors. This has often led to the Table 2 fallacy, where one presents the coefficient estimates for all covariates from a single multivariable regression model, which are often uninterpretable in a descriptive analysis. We argue that machine learning (ML) is a potential solution to this problem. We illustrate the power of ML with an example analysis identifying the most important predictors of alcohol abuse among sexual minority youth. The framework we propose for this analysis is as follows: (1) Identify a few ML methods for the analysis, (2) optimize the parameters using the whole data with a nested cross-validation approach, (3) rank the variables using variable importance scores, (4) present partial dependence plots (PDP) to illustrate the association between the important variables and the outcome, (5) and identify the strength of the interaction terms using the PDPs. We discuss the potential strengths and weaknesses of using ML methods for descriptive analysis and future directions for research. R codes to reproduce these analyses are provided, which we invite other researchers to use.

Shape memory alloys (SMAs) have found widespread application in various fields of science and technology due to their unique properties, such as superelasticity and shape memory effect. These alloys retain their initial form by memorising it between two transformation phases, which is temperature or magnetic field-dependent. The application of such materials is straightforward. The alloy can be deformed by force and recover to its initial shape or size after heating over a specific temperature. There are a lot of various kinds of SMA, for instance, Fe–Mn–Si, Cu–Zn–Al, and Cu–Al–N, and every type of SMA is applied specifically, though Nitinol Ni-Ti is ubiquitous because of its stable properties SMAs are widely used in medicine, the aerospace industry, motor building, civil engineering, dentistry, etc. During their operation, structural elements made of SMAs undergo long-term cyclic loading that can lead to premature loss of functional properties, exhaustion of lifetime, and subsequent failure. Therefore, ensuring sufficient functional properties and endurance of SMA is necessary. Often, the experiments are quite costly and time-consuming and require expert knowledge. Therefore, it is crucial to model the functional and structural properties of SMAs by employing AI (Artificial intelligence) and machine learning (ML) methods. AI can be employed to model SMA behaviour. AI is actively used in material science and fracture mechanics ML is a part of AI that can efficiently solve complicated tasks. This study aims to perform a comprehensive review of the application of ML methods to estimate various properties of shape memory alloys. A comprehensive analysis of ML methods was performed as applied to modelling various properties of SMAs. Several studies concern the application of methods of AI and ML to solve such problems. In general, AI and ML methods are promising and powerful tools to model the SMAs properties. Nevertheless, there is always room for improvement and further elaboration of the aforementioned methods and approaches for modelling the functional and structural properties of SMAs

This study conducts a comparative analysis of various machine learning methods for classifying the quality of Palu shallots based on the Indonesian National Standard (SNI). The dataset consists of 1,500 samples of Palu shallots, each characterized by 10 key features, including size, color, texture, and moisture content. Five machine learning models—Naïve Bayes, Decision Tree, Random Forest, Support Vector Machine (SVM), and Logistic Regression—were evaluated using accuracy, precision, recall, and F1 score as performance metrics. The results indicate that Random Forest achieved the best performance with an accuracy of 95.4%, followed by Decision Tree (90.7%) and SVM (90.2%). Random Forest also excelled in precision (93.6%) and F1 Score (93.5%), making it the most reliable model for shallot quality classification. Meanwhile, SVM demonstrated a good balance between recall and precision, making it a strong alternative. Implementing machine learning models has the potential to enhance the efficiency and accuracy of agricultural product quality assurance. The findings of this study provide valuable insights for farmers, agribusiness practitioners, and researchers adopting artificial intelligence technology for more precise and efficient agricultural quality assessment.

A survey on machine learning models for financial time series forecasting

Due to the exponential rise of mobile technology, a slew of new mobile security concerns has surfaced recently. To address the hazards connected with malware, many approaches have been developed. Signature-based detection is the most widely used approach for detecting Android malware. This approach has the disadvantage of being unable to identify unknown malware. As a result of this issue, machine learning (ML) for identifying and categorising malware apps was created. Conventional ML methods are concerned with increasing classification accuracy. However, the standard classification method performs poorly in recognising malware applications due to the unbalanced real-world datasets. In this study, an empirical analysis of the detection performance of ML methods in the presence of class imbalance is conducted. Specifically, eleven (11) ML methods with diverse computational complexities were investigated. Also, a synthetic minority oversampling technique (SMOTE) and random undersampling (RUS) are deployed to address the class imbalance in the Android malware datasets. The experimented ML methods are tested using the Malgenome and Drebin Android malware datasets that contain features gathered from both static and dynamic malware approaches. According to the experimental findings, the performance of each experimented ML method varies across the datasets. Moreover, the presence of class imbalance deteriorated the performance of the ML methods as their performances were amplified with the deployment of data sampling methods (SMOTE and RUS) used to alleviate the class imbalance problem. Besides, ML models with SMOTE technique are superior to other experimented methods. It is therefore recommended to address the inherent class imbalance problem in Android Malware detection.

This work deals with one of the important topics - the organization of safe and optimal movement of land transport vehicles. Organizing the right logistics leads to economic gain as well as a reduction of greenhouse gas emissions into the environment. The main difficulty in solving this problem is that the problem is multiparametric, and dynamically variable in time. Another problem is that because of the above features it is not possible to write an algorithm that will choose the optimal route at each moment of time due to the complexity and variability of the input parameters. An interesting task is the development of software allowing automatic collection of traffic information, and transmission to the input of an adaptive decision-making system for choosing the most correct route. It is therefore proposed to develop an adaptive system based on modern machine learning techniques and genetic algorithms. To meet these challenges, the authors developed machine learning and simulation approaches. The work includes an analysis of machine learning methods, especially the use of neural networks in reinforcement training, as well as an analysis of machine learning methods for the task of finding the optimum route of transport. As a result, software has been developed for the crucial automated transport task through machine learning methods, analysis of sustainability of transport solutions based on machine learning methods and analysis of machine learning supported by digital control systems.

Introduction. Machine learning methods and elements of artificial intelligence are used to analyze random data, processes and signals. The study of relevant tools is already included in the various levels curricula. The purpose of the study is to demonstrate, using examples available to students of various specialties, that the error analysis of machine learning methods in solving specific tasks can be the basis in the educational process for the skills formation of using artificial intelligence elements.Materials and Methods. For processing random signals and data, widely available software is used: Microsoft Excel for preparing training and test samples, the Deductor analytical platform for implementing machine learning algorithms. As an example, quasi-harmonic signals with random parameters are processed for technical specialties, and the results of psycho diagnostics are used to process multidimensional random data.Results. As a typical solution of approximation technical problems, direct propagation neural network errors in using to determine random signal parameters are analyzed. As a solution of classification problems, multidimensional random data with different dimensions were processed using neural networks and the "decision tree" method. The advantages of the combined use of these two machine learning methods are analyzed. These examples and their analysis were tested in classes with university students in the disciplines of "Digital Signal Processing" and "Fundamentals of Statistics".Discussion and Conclusions. The statistical features of the obtained results, the possibilities of reducing the training sample and selective analysis of multidimensional random data are discussed. It is shown that an adequate assessment of the machine learning methods errors can significantly expand the possibilities of their application, and can be the basis for the formation of skills for their use.

The history of using machine learning algorithms to analyze statistical models is quite long. The development of computer technology has given these algorithms a new breath. Nowadays deep learning is mainstream and most popular area in machine learning. However, the authors believe that many researchers are trying to use deep learning methods beyond their applicability. This happens because of the widespread availability of software systems that implement deep learning algorithms, and the apparent simplicity of research. All this motivate the authors to compare deep learning algorithms and classical machine learning algorithms. The Large Hadron Collider experiment is chosen for this task, because the authors are familiar with this scientific field, and also because the experiment data is open source. The article compares various machine learning algorithms in relation to the problem of recognizing the decay reaction + + + at the Large Hadron Collider. The authors use open source implementations of machine learning algorithms. We compare algorithms with each other based on calculated metrics. As a result of the research, we can conclude that all the considered machine learning methods are quite comparable with each other (taking into account the selected metrics), while different methods have different areas of applicability.

Accurately determining the favorable areas of geothermal resources and selecting the target positions of exploration wells are extremely important for exploration and efficient development. This study used the Pearson correlation coefficient and Gini gain to analyze five influencing factors related to the presence of economically viable geothermal potential. The evaluation model of the favorable areas was constructed by using different Machine Learning (ML) methods: Bayesian classifier (Bayes), Support Vector Machine, Bootstrap Aggregating (Bagging), BP neural network, Decision Tree and Logistic Regression classification. The quality of each model was verified by statistical evaluation indicators: Accuracy (ACC), F 1 score (F 1) and Receiver Operating Characteristic curve (ROC curve). The methodology was applied to the case study of Xinjiang Uygur Autonomous Region, China. Due to the results obtained, all ML models showed strong prediction and classification performance on the target area selection of geothermal exploration, as evidenced by each model’s metrics: the ACC was above 80%, the F 1 was above 0.8, and the Area Under the ROC Curve (AUC) was greater than 0.85. The metrics obtained by the Bagging method were the highest. Finally, the results of the six ML models were combined to classify the study area’s geothermal potential, which was consistent with the available information. This study provides a specific basis and technical support for applying the method in further surveys and campaigns.

Machine learning techniques can extensively apply in the solution of the medicine domain problems by applying classification models and systems that can support medical personnel in the diagnosis and predication of diagnosis diseases. Though, it's hard to extract knowledge and information from medical records and data because this data and information is in mixed, unorganized, and high dimensional. This data also contains noise in collected data and outliers exist in collected data. Main applicable method will be used applies by checking different machine learning techniques. The performance of machine learning technique is checked by verifying and validating machine learning techniques' performances through accuracy. Present research paper has been discussing about the usability and applicability of different machine learning techniques i.e. decision tree algorithm, support vector machine method, random forest method, evolutionary algorithms based models and swarm intelligence based techniques in the diagnosis and treatment of the diseases. Advance medical diagnosis criteria generates confidence in diagnosis by using imagining techniques in the diagnosis of a disease is extensively used by doctors. In view of the fact that analyzing medical images is very complex and difficult task, by using machine learning methods for analysis of imaging will support and give major help in disease diagnosis. Application of different Machine learning methods is used by applying its techniques on big data for interpretation for diagnosis because machine learning methods show their capability and shows their easiness to solve the problems of bioinformatics domain.

The efficacy of ethylene-vinyl alcohol copolymer films (EVOH) incorporating the essential oil components cinnamaldehyde (CINHO), citral (CIT), isoeugenol (IEG), or linalool (LIN) to control growth rate (GR) and production of T-2 and HT-2 toxins by Fusarium sporotrichioides cultured on oat grains under different temperature (28, 20, and 15 °C) and water activity (aw) (0.99 and 0.96) regimes was assayed. GR in controls/treatments usually increased with increasing temperature, regardless of aw, but no significant differences concerning aw were found. Toxin production decreased with increasing temperature. The effectiveness of films to control fungal GR and toxin production was as follows: EVOH-CIT > EVOH-CINHO > EVOH-IEG > EVOH-LIN. With few exceptions, effective doses of EVOH-CIT, EVOH-CINHO, and EVOH-IEG films to reduce/inhibit GR by 50%, 90%, and 100% (ED50, ED90, and ED100) ranged from 515 to 3330 µg/culture in Petri dish (25 g oat grains) depending on film type, aw, and temperature. ED90 and ED100 of EVOH-LIN were >3330 µg/fungal culture. The potential of several machine learning (ML) methods to predict F. sporotrichioides GR and T-2 and HT-2 toxin production under the assayed conditions was comparatively analyzed. XGBoost and random forest attained the best performance, support vector machine and neural network ranked third or fourth depending on the output, while multiple linear regression proved to be the worst.

This paper introduces a new image-based handwritten historical digit dataset named Arkiv Digital Sweden (ARDIS). The images in ARDIS dataset are extracted from 15,000 Swedish church records which were written by different priests with various handwriting styles in the nineteenth and twentieth centuries. The constructed dataset consists of three single-digit datasets and one-digit string dataset. The digit string dataset includes 10,000 samples in red–green–blue color space, whereas the other datasets contain 7600 single-digit images in different color spaces. An extensive analysis of machine learning methods on several digit datasets is carried out. Additionally, correlation between ARDIS and existing digit datasets Modified National Institute of Standards and Technology (MNIST) and US Postal Service (USPS) is investigated. Experimental results show that machine learning algorithms, including deep learning methods, provide low recognition accuracy as they face difficulties when trained on existing datasets and tested on ARDIS dataset. Accordingly, convolutional neural network trained on MNIST and USPS and tested on ARDIS provide the highest accuracies 58.80% and 35.44%, respectively. Consequently, the results reveal that machine learning methods trained on existing datasets can have difficulties to recognize digits effectively on our dataset which proves that ARDIS dataset has unique characteristics. This dataset is publicly available for the research community to further advance handwritten digit recognition algorithms.

A systematic literature review of machine learning methods applied to predictive maintenance

The article is devoted to the application of machine learning methods in predicting the formation of prospective sectors of the new generation economy. In the conditions of modern digital transformations, it has been shown that the replacement of the traditional existing economy with new generation economic models is one of the world's priority development directions. The relevance of the application of Machine learning (ML) methods, one of the artificial intelligence (AI) technologies, in improving the processes of formation and development of the traditional sectors of the economy, and in forecasting its new generation perspective sectors is substantiated. An analysis of scientific research works related to the problem was carried out. Digital transformation and technologies, sustainability and sustainability, greening technologies and circularity, joint use, smart decision-making and management, platforms and ecosystems, innovative entrepreneurship, research and economic development, inclusion, and social development, Industry 5.0 platform technologies of the formation of the new generation technological economy. the main basic principles such as transition etc. have been worked out and the problems of its formation have been analyzed. The features and perspectives of the application of machine learning methods in forecasting the prospective sectors of the new generation economy have been explained. Classification features of machine learning methods are explained and its models are shown. The structural scheme of the stages of economic development forecasting has been developed and information about its methods has been provided. A comparative analysis of machine learning methods applied in forecasting was carried out. The structural scheme of the application stages of the machine learning method in the forecasting process has been developed. Relevant recommendations on the application of Industry 4.0 platform technologies were given for forecasting the formation of prospective sectors of the new generation economy based on real data.

Inference of mechanical properties and structural grades of bamboo by machine learning methods