Search

Parkinson’s Disease (PD) diagnosis with an explicit severity level assessment is essential for timely medical intervention. Gait analysis has been proven to assist in defining PD progression and Machine Learning algorithms aid in predicting the PD severity by learning intricate details from multidimensional gait datasets. Classifiers that best predict the severity level with better explainability of results are thus a requisite. This paper proposes an Information Set-based Decision Tree (IFS-DT), a classification algorithm that adopts a probabilistic-possibilistic approach for predicting the target variable. Information carried by each feature of the dataset is utilized rather than the possibility of its occurrence which reduces uncertainty in determining the outcome. PD gait dataset built from VGRF signals obtained from the PhysioNet database is utilized for analyzing the proposed work. The IFS-DT classifier performs better than conventional DTs and provides results comparable to that obtained by an ensemble of classifiers. Statistical analyses such as the Wilcoxon signed-rank test are conducted to validate the significance of the findings. Moreover, the importance of training a model with a balanced dataset with optimal sample size for obtaining reliable prediction results is also discussed.

Density-based spatial clustering of noisy applications (DBSCAN), a widely used density-based clustering technique, faces challenges in determining its key parameter, Eps, leading to manual specification and suboptimal clustering outcomes. Additionally, its time complexity poses limitations. This study introduces a novel approach to enhance the algorithm’s performance. We combined the DBSCAN algorithm with another approach, the particle swarm optimization, based on a novel way to represent text, termed a dependency graph; this method is recognized as DGBPSO-DBSCAN (Dependency Graph Based Particle swarm algorithm for DBSCAN). This paper focuses on employing a novel approach for PSO variable upgrading to explore the Eps range more rapidly and effectively. This method offers a selection of informative elements from the text, where the initial groups are derived from the graph-based degree centrality, the PSO method is used to choose the most compelling features, and the DBSCAN algorithm is used for text clustering. The experimental findings indicate that the modified PSO is used to improve DBSCAN. We compared the outcomes of the suggested technique to those of the standard clustering algorithm. According to the assessment criteria MSE, accuracy, precision, recall, and F-measure, our technique has been demonstrated to be superior to the conventional method.

Rife acceptance of Cloud Computing has made it bull’s eye for the hackers. Intrusion detection System (IDS) plays a vibrant role for it. Researchers have done marvelous works on the development of a competence IDS. But there are many challenges still exists with IDS. One of the biggest concerns is that the computational complexity and false alarms of the IDS escalates with the increase in the number of features or attributes of the dataset. Hence, the concept of Feature Selection (FS) contributes an all-important role for the buildout of an efficacious IDS. New FS algorithm is put forward which is the modified Firefly Algorithm in which Decision Tree (DT) classifier is used as the classification function. We have used the hybrid classifier which is the combination of neural network and DT. We have used CSE CIC IDS 2018 dataset and simulated dataset for performance assessment. Our examination pragmatic that the performance of proposed architecture is better than the state-of-the-art algorithms.

Machine learning (ML) has been instrumental in solving complex problems and significantly advancing different areas of our lives. Decision tree-based methods have gained significant popularity among the diverse range of ML algorithms due to their simplicity and interpretability. This paper presents a comprehensive overview of decision trees, including the core concepts, algorithms, applications, their early development to the recent high-performing ensemble algorithms and their mathematical and algorithmic representations, which are lacking in the literature and will be beneficial to ML researchers and industry experts. Some of the algorithms include classification and regression tree (CART), Iterative Dichotomiser 3 (ID3), C4.5, C5.0, Chi-squared Automatic Interaction Detection (CHAID), conditional inference trees, and other tree-based ensemble algorithms, such as random forest, gradient-boosted decision trees, and rotation forest. Their utilisation in recent literature is also discussed, focusing on applications in medical diagnosis and fraud detection.

An increase in vehicular traffic and a scarcity of parking spaces are creating significant challenges for urban parking management. This study aims to tackle these issues that escalate congestion and pollution and decrease urban productivity, by utilizing machine learning models to accurately predict parking space availability and categorize occupancy levels. It employs a dataset from a college campus garage collected from January 2022 to June 2023 and analyzes the performance of random forest, decision tree, linear regression, and support vector models by comparing them, using multiple evaluation metrics. The results revealed that the random forest model was the most reliable, as it demonstrated strong performance in both the regression and classification analyses and was adept at estimating the exact number of available parking spaces. A concurrent classification analysis that categorized parking occupancy into different levels proved valuable for enhancing the quality of communication and decision-making. An analysis of the importance of various features clearly highlighted the influence of the day of the week on parking demand and patterns; the impact of seasonality on the volume of parking usage; and the time of day, which plays a crucial role in determining parking behavior. The research will benefit urban planners, facility managers, and policymakers by providing them with insights and tools that will enhance the urban parking experience and address the complex challenges of modern urban environments.

Cardiovascular problems have emerged as a significant concern, adversely impacting individuals across all age groups. Several recent research studies have used Machine learning (ML) techniques to design decision-making systems for the tremendous data in the medical sector. Although these works obtained promising results, most of the studies focused on small datasets. Since the size of the dataset affects algorithm performance, this study used two datasets, such as Kaggle’s heart disease dataset of over 70,000 records and UCI’s heart disease dataset of 1025 records. In addition to the old features the Pulse Pressure (PP), the Body Mass Index (BMI), and the Mean Arterial Pressure (MAP), three new features are introduced to improve the results. This paper proposes the TLV (Two-Layer Voting) model, which is an ensemble method of hard and soft voting. As part of layer 1, features are shortlisted by soft and hard voting using three statistical methods, including the ANOVA f-test, Chi-squared test, and Mutual Information. In layer 2, soft voting and hard voting performance are compared, which incorporates Multi-Layer Perceptron, Decision Tree, Support Vector Classifier, and Random Forest algorithms. Classification algorithms are hyper-tuned using the GridSearchCV method in the second layer. Using UCI’s heart disease dataset and Kaggle’s CVD dataset, the proposed TLV methodology with soft voting provided the highest accuracy of 99.03% and 88.09%, respectively. The proposed model significantly outperforms existing CAD disease prediction studies.

This study explores the implementation of lightweight binary classification algorithms on low-cost Field-Programmable Gate Arrays (FPGAs) for medical image analysis. Recognizing the growing demand for efficient and accurate diagnostic tools in healthcare, we focus on applying FPGAs to process small-sized medical images, explicitly targeting the detection of malaria from blood cell images. Our approach involves the hardware designs of k-nearest Neighbors (k-NN), Convolutional Neural Networks (CNN), and Decision Tree classifiers rigorously tested on a publicly available Malaria dataset. The methodology emphasizes the integration of these classifiers on FPGA, detailing the optimization strategies that allow for enhanced processing speed and reduced resource utilization. Comparative analysis reveals that our FPGA-based implementation significantly outperforms MATLAB simulations, achieving processing speeds more than thousands of times faster. Our proposed hardware design also requires fewer Look-up Tables (LUTs) than other classification studies in the literature, showcasing decreases of 73.9% for k-NN, 57% for CNN, and 96.7% for the Decision Tree classifier. Furthermore, results highlight the Decision Tree classifier as the most effective, with an accuracy rate of 99.33%, followed by CNN at 97.67% and k-NN at 95.33%. These findings demonstrate the capability of FPGAs in medical image classification and underscore their potential to revolutionize disease diagnosis processes, particularly in resource-limited environments.

The manufacturing of hard disk drives involves the intricate assembly of numerous components, making the testing process time-consuming and resource intensive. To optimize the manufacturing process and increase testing efficiency, the development of a rule-based expert system is proposed. This system leverages predictive models constructed from assembly process data to identify potentially defective hard drives before undergoing extensive testing. By preemptively identifying defects, this approach substantially reduces testing time and enhances tester capacity. Given the categorical and imbalanced nature of assembly data, Decision Trees are employed as the prediction model. Specifically, three Decision Tree algorithms are explored: ID3, C4.5, and CART. In addition, four feature selection techniques, namely Information Gain, Gain Ratio, Chi-Square, and Symmetrical Uncertainty, are utilized to identify high-impact features. Our experimental findings reveal that Information Gain coupled with the C4.5 algorithm yields the most favorable results in terms of prediction accuracy, modeling efficiency, and rule generation. Moreover, our study establishes that setting the failure probability threshold between 0.15 and 0.70 provides the shortest total test time for the proposed process, as supported by a 95% confidence level. This achievement represents a statistically significant enhancement compared with the existing manufacturing process.

This work presents a metamaterial absorber (MMA) for X- and Ku-bands with a metallic resonating patch on top and a ground plane separated by substrate FR-4 with a thickness of $0.053~\lambda $ at the lowest resonance frequency. The proposed MMA demonstrates perfect absorption of 99.42, 98.48, 98.92, and 99.34 % at 9.948, 13.26, 14.92, and 15.80 GHz, respectively at normal incidence. The proposed MMA demonstrates perfect absorption for a polarization and incident angle over a wide range of angles up to 45°. To understand the fundamental EM behavior of the metamaterial structure, equivalent circuit analysis was carried out, and the circuit outputs accorded with the simulation results. This article also compares various machine learning (ML) methods for optimizing the design and predictive modeling of MMAs, such as decision trees, K-nearest neighbors, random forests, extra trees (ET), bagging, LightGBM, XGBoost, hist gradient boosting, cat boost, and gradient boosting regressors. The primary objective is to assess the usefulness of each regressor technique in estimating the performance of MMAs using multiple tests ranging from TC-40 to TC-80 and performance metrics such as adjusted R-squared score, MSE, RMSE, and MAE, in which the ET regressor excels. Simulation results suggest that ML-based techniques can save simulation resources and time while still being an efficient tool for predicting absorber behavior at intermediate and subsequent frequencies.

The world has embarked on a road to sustainable energy production. As a result, countries have turned to microgrid developments. This article aims to study the feasibility of renewable sources such as solar PV and wind power for integrating a microgrid campus, taking the example of a case in East Africa, precisely the case of the University of Djibouti. We applied the weather parameters to evaluate the solar and wind potential with the Decision Tree method for analyzing and classifying the degrees of solar radiation and the consistency of wind speed. These data are spread over eight years to establish and capture seasonal changes and prove the accessibility of renewable sources in a specific site. The results were compared to Random Forest, Logistic Regression, K-Nearest Neighbors, Support Vector Machine, and Naïve Bayes classifiers, which showed that the performance of classifying the Decision tree outperformed all other methods with an accuracy of 0.99321. The second work of this article explored the forecasting of the possible powers predicted with the LSTM deep learning method by the generation of the Solar PV array and wind turbines which were simulated on PVLib and Windpowerlib. The results are favorable, and the LSTM has performed well on the different hyperparameters. With the combination of machine learning and deep learning, it was possible to theoretically conclude the integration of renewable energies since we investigated all the potential possibilities in evaluating meteorological parameters and power predictions. Finally, decision scores from the Decision Tree architecture and the LSTM features were integrated to form a hybrid Tree-LSTM fusion method. It introduces a novel architectural concept that can effectively address sequential data and harness the non-linear capabilities of decision trees. The proposed model was validated by tuning the hyperparameters. Enhancing the maximum depth of the model increases the performance at a certain point, and conversely, reducing the minimum sample split improves the model performance. These contributions will help to create sustainable energy systems and increase the transition to a clean CO2 environment.

This research paper presents an approach for recognizing motor imagery (MI) movements through brain signals, which has essential applications in assisting people with mobility disorders. One of the critical challenges in this field is that such individuals should be exposed to their surroundings with the help of exact motion recognition. This article represents an algorithm where Fourier-based filters are used for obtaining sub-bands of EEG signals for motion recognition and brain computer interface (BCI) application. Specifically, we segment motor imagery signals into eight orthogonal Fourier intrinsic band functions (FIBFs) and extract statistical feature matrices from each FIBF. We then propose a methodology derived from the k-nearest neighbor (kNN) classifier which is also compared with state-of-the-art classifiers like decision tree (DT), support vector machine (SVM), and naive Bayes (NB), to classify the extracted features and to establish its outperforming nature. Our experimental results show that our proposed approach achieves the highest classification accuracy of 96% and 84.03% with the kNN classifier on the BCI III IVa and BCI IV 2a datasets, outperforming state-of-the-art methods. These results demonstrate the potential of our approach in enabling accurate motion recognition and assisting people with mobility disorders.

Understanding the factors contributing to depression in farmers is crucial for ensuring their well-being and productivity. To address this issue, our study delves into depression factors among farmers, employing advanced tree-based machine learning (ML) algorithms, specifically focusing on the Category Boosting (CatB) algorithm. Applying the Patient Health Questionnaire-9 (PHQ-9) criteria, 2,446 individuals among 14,810 repondents were classified into depression including mild symptoms. In the classification, CatB achieved an impressive 79.7% accuracy and 81.4% F1 score compared to the other tree-based ensemble models (Random Forest - RF, Extra Trees - ET, and XGBoost - XGB). RF showed the highest sensitivity at 90.0% and the 81.3% F1 score followed by CatB. For the feature importances, the Gini impurity was predominantly used to assess in the RF and ET models. Through the analysis of feature importances, ‘Health’, ‘Sleep time’, ‘Busyness’, ‘Income’, and ‘Frequency of wearing protective gear’ were identified as significant features. These results highlighted the significance of treatment strategies for individuals at high risk. and developing treatment strategies for high-risk individuals in the agricultural sector. Empowering healthcare providers by giving them access to this tool can lead to more effective interventions, potentially reducing the burden of depression and enhancing farmers’ productivity.

Agriculture contributes a significant amount to the economy of India due to the dependence on human beings for their survival. The main obstacle to food security is population expansion leading to rising demand for food. Farmers must produce more on the same land to boost the supply. Through crop yield prediction, technology can assist farmers in producing more. This paper’s primary goal is to predict crop yield utilizing the variables of rainfall, crop, meteorological conditions, area, production, and yield that have posed a serious threat to the long-term viability of agriculture. Crop yield prediction is a decision-support tool that uses machine learning and deep learning that can be used to make decisions about which crops to produce and what to do in the crop’s growing season. It can decide which crops to produce and what to do in the crop’s growing season. Regardless of the distracting environment, machine learning and deep learning algorithms are utilized in crop selection to reduce agricultural yield output losses. To estimate the agricultural yield, machine learning techniques: decision tree, random forest, and XGBoost regression; deep learning techniques - convolutional neural network and long-short term memory network have been used. Accuracy, root mean square error, mean square error, mean absolute error, standard deviation, and losses are compared. Other machine learning and deep learning methods fall short compared to the random forest and convolutional neural network. The random forest has a maximum accuracy of 98.96%, mean absolute error of 1.97, root mean square error of 2.45, and standard deviation of 1.23. The convolutional neural network has been evaluated with a minimum loss of 0.00060. Consequently, a model is developed that, compared to other algorithms, predicts the yield quite well. The findings are then analyzed using the root mean square error metric to understand better how the model’s errors compare to those of the other methods.

Power quality disturbances (PQDs) in modern electrical power systems, caused by the integration of nonlinear power electronic devices and erratic distributed generation (DG), lead to interruptions and significant energy losses for end users. However, conventional methods for PQDs classification face challenges in dealing with noise interference and feature selection. To address these challenges, this research paper proposes a novel approach that combines the stockwell transform (ST) with an improved grey wolf optimization-based kernel extreme learning machine to enhance classification accuracy. The stockwell transform is utilized to extract meaningful features from the power quality (PQ) signals, which are subsequently input into the kernel extreme learning machine (KELM). Furthermore, the parameters of the KELM model are optimized using the improved grey wolf optimization (IGWO) approach to improve the accuracy of classification. To evaluate the performance of the proposed method, real-time implementation is considered by incorporating PQDs data with signal-to-noise ratios (SNR) of 20 dB, 30 dB, and 40 dB into the original synthetic signals. Multiple noise conditions are simulated to assess the proposed model ability to identify and classify disturbance signals. The results demonstrate a detection accuracy of 99.76% under noiseless conditions, indicating the model’s high accuracy. Moreover, the proposed method exhibits robustness to noise, achieving accuracies of 98.86%, 98.32%, and 97.3% at SNR levels of 40 dB, 30 dB, and 20 dB, respectively. In the end, this work has performed a comparative study with other previously published work. Compared with other classification methods, the algorithm proposed in this paper has higher accuracy, and it is an efficient and feasible classification method.

Non-intrusive load monitoring (NILM) detects the energy consumption of individual appliances by monitoring the overall electricity usage in a building. By analyzing voltage and current characteristics, NILM can recognize the usage patterns of various appliances, thus facilitating energy conservation and management. To implement non-intrusive load classification and recognition more effectively, this study proposes an ensemble learning algorithm based on soft voting, which comprises a decision tree, K-nearest neighbor algorithm, and multilayer perceptron (EL-SV $_{\mathrm {DT-KNN-MLP}}$ ). In this study, the voltage and current features in the plug-load appliance identification dataset (PLAID) and worldwide household and industry transient energy dataset (WHITED) are used as input data. The dataset is examined thoroughly and preprocessed before it is fed into the EL-SV $_{\mathrm {DT-KNN-MLP}}$ . During preprocessing, six different normalization techniques are applied to the data to improve the accuracy and reliability of the machine-learning model, thus rendering the proposed algorithm more adept at classifying and recognizing appliances. The proposed method is validated by comparing it with other machine learning algorithms in terms of accuracy, precision, recall, and F1 score under the six different normalization methods. For the PLAID, the proposed algorithm can achieve high accuracy scores of 99.79%, 98%, 98.11%, 98.36%, 96.42%, and 98.76% under the min–max normalization, MaxAbs scaler, robust scaler, z-score normalization, L1 normalization, and Yeo–Johnson transformation, respectively. Similarly, for the WHITED, the proposed algorithm can achieve high accuracy scores of 99.31%, 98.14%, 98.3%, 98.35%, 97.65%, and 98.02% under the abovementioned normalization methods. The results show that the proposed EL-SV $_{\mathrm {DT-KNN-MLP}}$ algorithm outperforms the other ten machine learning algorithms examined in this study.

The Internet of Things (IoT) is a form of Internet-based distributed computing that allows devices and their services to interact and execute tasks for each other. Consequently, the footprint of the IoT is increasing and becoming more complex to the highest degree. This has also given birth to new IoT-enabled applications and services. Efficient service interaction and management also call for understanding and analyzing the nature of IoT services. Further, IoT services must be characterized into various classes, and different service-related attributes must be considered for the classification. This article assesses the requirements of heterogeneous IoT services by examining their interactions. Principally, heterogeneous IoT and their service interactions are targeted. The research work performs classification of IoT services into various classes. Services are classified on the basis of various attributes. The attributes reflect different characteristics of the services. This research enables improved utilization of IoT services through efficient classification of available resources using machine learning methods. To demonstrate service classification applicability, the SVM, voting classifier, and decision tree have been applied in a service-oriented environment along with different types of services. All the services in the data set were analyzed and divided into five classes. Moreover, the decision tree performed well and achieved higher accuracy values in all classes. However, the overall prediction and classification of the decision tree model were observed to be good and satisfactorily high.

Cyber-attacks pose increasing challenges in precisely detecting intrusions, risking data confidentiality, integrity, and availability. This review paper presents recent IDS taxonomy, a comprehensive review of intrusion detection techniques, and commonly used datasets for evaluation. It discusses evasion techniques employed by attackers and the challenges in combating them to enhance network security. Researchers strive to improve IDS by accurately detecting intruders, reducing false positives, and identifying new threats. Machine learning (ML) and deep learning (DL) techniques are adopted in IDS systems, showing potential in efficiently detecting intruders across networks. The paper explores the latest trends and advancements in ML and DL-based network intrusion detection systems (NIDS), including methodology, evaluation metrics, and dataset selection. It emphasizes research obstacles and proposes a future research model to address weaknesses in the methodologies. The decision tree, known for its speed and user-friendliness, is proposed as a model for detecting result anomalies, combining findings from a comparative survey. This research aims to provide insights into building an effective decision tree-based detection framework.

Packet classification is the core technology that underpins software-defined networking (SDN). With the rapid development of network applications, the increasing complexity of flow tables in SDN brings challenges to update and classification time. In order to achieve fast search and update at the same time, this article proposes a high-performance packet classification method PcmSU based on decision tree. The method optimizes the search and update process in the traditional decision tree-based classification method, and the classification efficiency of the constructed decision tree has been greatly improved compared with that before optimization. Based on the characteristics of the small interval, the reconstruction of the decision tree is greatly reduced and the rule update efficiency is improved. In addition, based on the characteristics of the decision tree constructed by this method, there is no need to consider the specific location and sequence number of rule updates when adding rules, which can avoid introducing rule conflicts. Experimental results show that the PcmSU method not only supports high-speed packet classification and linear memory consumption, but also has fast rule update speed.

Network intrusion detection is an important technology in national cyberspace security strategy and has become a research hotspot in various cyberspace security issues in recent years. The development of effective and efficient intelligent network intrusion detection methods using advanced machine learning algorithms is of great importance for defending against various network intrusions in complex network environments. In this study, a network intrusion detection method based on decision tree twin support vector machine and hierarchical clustering, named HC-DTTWSVM, is proposed, which can effectively detect different categories of network intrusion. First, the hierarchical clustering algorithm is applied to construct the decision tree for network traffic data, where the bottom-up merging approach is used to maximize the separation of the upper nodes of the decision tree, which reduces the error accumulation in the construction of the decision tree. Then, twin support vector machines are embedded in the constructed decision tree to implement the network intrusion detection model, which can effectively detect the network intrusion category in a top-down manner. The detection performance of the proposed HC-DTTWSVM method is evaluated on NSL-KDD and UNSW-NB15 intrusion detection benchmark datasets. Experimental results show that HC-DTTWSVM can effectively detect different categories of network intrusion and achieves comparable detection performance compared to some of the recently proposed network intrusion detection methods.

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 $\mathcal {Q}$ -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.

In the present, we can use more than 200 different methods and algorithms for automatic recognition systems, the basis of which is the recognition of a face in an image, subsequent extraction of areas of interest and classification of individual parameters using neural networks or other classifications. Currently, one of the most discussed research topics connecting the fields of psychology and artificial intelligence is the classification of emotions from behavioral characteristics. We created the Emotnizer application for collecting and processing behavioral characteristics. The input characteristics record 24 different ways of user behavior when working with a mouse and keyboard, e.g., when rewriting text (determining the method of clicking / key pressing speed, cursor changes, errors in the text, etc.). In the article, we analyze the obtained parameters using the decision tree method and find out whether it is possible to successfully classify the emotional state using these parameters. We found that for the successful classification of emotional states using behavioral characteristics, it is appropriate to classify mainly emotional states with a high value of valence and arousal.

Packet classification is the core technology of network layer and an important means to ensure the security of network system. With the rapid development of network technology, higher requirements are put forward for the speed of network packet classification. This paper improves the traditional single thread package classification framework, A new parallelization method for fast packet classification (MpFPC) based on distributed computing is proposed, the method adopts the packet classification idea based on decision tree, but compared with the traditional algorithm, a rule mapping preprocessing process is added before constructing the classification decision tree, which effectively removes the rule redundancy and conflict, so as to avoid the rule replication problem of the traditional decision-tree-based method. In addition, the method can group the rules and data packets at the same time, which improves the packet classification efficiency. Experimental results show that MpFPC method has high classification efficiency and has obvious speed advantage compared with Uscuts method with time complexity of $O(k$ log $n$ ). In addition, the test results also show that the classification speed of MpFPC will increase with the increasing number of computing nodes, which provides a new possible way to meet the classification wire-speed requirement.

Decision tree algorithm (DT) is a commonly used data mining method for classification and regression. DT repeatedly divides a dataset into pure subsets based on impurity measurements such as entropy and Gini. Then relatively “pure” partitions consisting of observations with the (almost) same class are obtained. Gini index is one of the representative indices for measuring the impurity of data. However, the Gini index does not take into account distances between classes. If the distances between classes are considered when measuring impurity, the decision tree algorithm can distinguish clearly observations with different classes. To the end, a new decision tree algorithm based on Rao-Stirling index is proposed considering distances between classes. Rao-Stirling index considers distances between classes in such a way that weights more to pairs of references in more distant classes when measuring data impurity. Experimental results indicate that the proposed method is superior in terms of accuracy, implying that considering the distances between classes can help improve accuracy in DT.

The power of mobile devices has been harnessed as a platform for augmented reality (AR) that has embarked on opportunities to combine more inputs called multimodal inputs. There has been little research into the multimodal inputs that combine emotion, speech and markers in a mobile AR learning environment. This study aims to propose a framework for a mobile AR learning system that incorporates the combination of multimodal inputs, namely emotion, image-based marker and speech in order to determine how such a combination can enhance the learning experience. The proposed framework integrates the multimodal inputs based on the decision tree and develops into a four-phase learning system based on Kolb’s experiential learning model. To evaluate this learning system, 38 students were selected and divided into two groups for an experiment on vocabulary at a primary school. Quantitative findings showed better results concerning learning effectiveness, mental load, engagement, competency and challenge when the three multimodal inputs, speech, marker and emotion, were combined. Therefore, the proposed multimodal framework can be used as a guideline to develop a multimodal based AR application for learning environment by integrating multimodal inputs and Kolb’s experiential learning model.

Machine learning and deep learning have gained a lot of attention from researchers because of their promising predictive performance and the availability of extensive high-dimensional data and high-performance computational hardware. However, the performance of these algorithms is susceptible to the choice of hyperparameters, while optimizing these hyperparameters is usually computationally expensive. For this reason, this study proposed a novel sequential search algorithm using decision tree regression as the surrogate function. In each iteration of this algorithm, several new combinations of hyperparameters were selected from a few best-performed leaves of the decision tree, called Iterative Decision Tree (IDT). Our approach could reduce the computational time from repetitive training the surrogate function compared to conventional sequential search algorithms and enable parallel computing. To confirm the effectiveness of the proposed algorithm, it was compared with six popular benchmark optimization algorithms, including Grid Search, Random Search, Bayesian Optimization, Random Forest, Tree-Structured Parzen Estimation, and Genetic Algorithm. The comparison was examined by optimizing three benchmark nonconvex functions, and hyperparameter tuning of two machine learning algorithms (Support Vector Machine and Random Forest) and two deep learning models (Autoencoder and Convolutional Neural Networks). As a result, the proposed algorithm achieved competitive performance with high stability in addition to the feature importance metrics.

Privacy protection has attracted increasing attention, and privacy concerns often prevent flexible data utilization. In most industries, data are distributed across multiple organizations due to privacy concerns. Federated learning (FL), which enables cross-organizational machine learning by communicating statistical information, is a state-of-the-art technology that is used to solve this problem. However, for gradient boosting decision tree (GBDT) in FL, balancing communication efficiency and security while maintaining sufficient accuracy remains an unresolved problem. In this paper, we propose an FL scheme for GBDT, i.e., efficient FL for GBDT (eFL-Boost), which minimizes accuracy loss, communication costs, and information leakage. The proposed scheme focuses on appropriate allocation of local computation (performed individually by each organization) and global computation (performed cooperatively by all organizations) when updating a model. It is known that tree structures incur high communication costs for global computation, whereas leaf weights do not require such costs and are expected to contribute relatively more to accuracy. Thus, in the proposed eFL-Boost, a tree structure is determined locally at one of the organizations, and leaf weights are calculated globally by aggregating the local gradients of all organizations. Specifically, eFL-Boost requires only three communications per update, and only statistical information that has low privacy risk is leaked to other organizations. Through performance evaluation on public data sets (ROC AUC, Log loss, and F1-score are used as metrics), the proposed eFL-Boost outperforms existing schemes that incur low communication costs and was comparable to a scheme that offers no privacy protection.

A combined cycle power plant (CCPP) employs gas and steam turbines to generate 50% more power while utilizing the same fuel as a normal single cycle plant. The performance of a CCPP under full load is affected by a variety of factors such as weather, process interactions, and coupling, which makes it challenging to operate. Therefore, a reliable assessment of the maximum output power of a CCPP is required to improve plant reliability and monetary performance. In this paper, a predictive model based on a generalized additive model (GAM) is proposed for the electrical power prediction of a CCPP at full load. In GAM, a boosted tree and gradient boosting algorithm are considered as shape function and learning technique for modeling a non-linear relationship between input and output attributes. Furthermore, predictive models based on linear regression (LR), Gaussian process regression (GPR), multilayer perceptron neural network (MLP), support vector regression (SVR), decision tree (DT), and bootstrap-aggregated tree (BBT) are also designed for comparison purposes. Results reveal that GAM improves the RMSE by 74%, 68.8%, 70.3%, 54.8%, 21.2%, and 17.3% compared to LR, GPR, MLP, SVR, DT, and BBT, respectively. Furthermore, the results of the Man-Whitney U test and rank analysis also confirm the effectiveness of GAM for energy prediction of CCPP. Finally, it can be concluded that the proposed method is effective, robust, and accurate for the assessment of the maximum output power of a CCPP to improve plant consistency and financial performance.

The energy management of buildings plays a vital role in the energy sector. With that in mind, and targeting an accurate forecast of electricity consumption, in the present paper is aimed to provide decision on the best prediction algorithm for each context. It may also increase energy usage related with renewables. In this way, the identification of different contexts is an advantage that may improve prediction accuracy. This paper proposes an innovative approach where a decision tree is used to identify different contexts in energy patterns. One week of five-minutes data sampling is used to test the proposed methodology. Each context is evaluated with a decision criterion based on reinforcement learning to find the best suitable forecasting algorithm. Two forecasting models are approached in this paper, based on K-Nearest Neighbor and Artificial Neural Networks, to illustrate the application of the proposed methodology. The reinforcement learning criterion consists of using the Multiarmed Bandit algorithm. The obtained results validate the adequacy of the proposed methodology in two case-studies: building; and industry.

With the rapid increase in the world’s population, the global electricity demand has increased drastically. Therefore, it is required to adopt efficient energy management mechanisms. Since the energy consumption trends are rather dynamic. Therefore, precise energy demand estimation and short and/or long-term forecasting results with higher accuracy are required to develop the optimization and control mechanism. Consequently, the machine learning (ML) techniques along with distributed demand response programs are being adopted to predict the future energy demand requirement with satisfactory results. In this paper, different state-of-the-art ML algorithms such as logistic regression (LR), support vector machines (SVM), naive Bayes (NB), decision tree classifier (DTC), K-nearest neighbor (KNN), and neural networks (NNs), have been implemented to analyze their performance. The main objective of this paper is to present a comparative analysis of ML algorithms for short-term load forecasting (STLF) regarding accuracy and forecast error. Based on the implementation and analysis, we have identified that, among other algorithms, the DTC provides comparatively better results. Therefore, we devised the enhanced DTC (EDTC) by integrating fitting function, loss function, and gradient boosting in DTC mathematical model for fine-tuning the control variables. The implementation results show that the proposed EDTC algorithm provides better forecast results (i.e., 99.9 % recall, 100% F1, 100% precision, 99.21 % training accuracy, and 99.70% testing accuracy.)

In recent years, Cloud computing has been developed and become the foundation of a wide range of applications. It allows users to access a catalog of standardized services and respond to their business needs flexibly and adaptively, in the event of unforeseen demands, paying solely for the consumption they have made. Task scheduling problem is considered one of the most critical cloud computing challenges. The problem refers to how to reasonably order and allocate the applications tasks provided by the users to be executed on virtual machines. Furthermore, the quality of scheduling performance has a direct effect on customer satisfaction. The task scheduling problem in cloud computing must be more accurately described in order to improve scheduling performance. In this paper, a multi-objective task scheduling algorithm is proposed based on the decision tree in a heterogenous environment. We introduce a new Task Scheduling-Decision Tree (TS-DT) algorithm for allocating and executing an application’s task. To evaluate the performance of the proposed TS-DT algorithm, a comparative study was conducted among the existing algorithms; Heterogeneous Earliest Finish Time (HEFT), Technique for Order of Preference by Similarity to Ideal Solution that incorporates the Entropy Weight Method (TOPSIS-EWM), and combining Q-Learning with the Heterogeneous Earliest Finish Time (QL-HEFT). Our results show that the proposed TS-DT algorithm outperforms the existing HEFT, TOPSIS-EWM, and QL-HEFT algorithms by reducing makespan by 5.21%, 2.54%, and 3.32%, respectively, improving resource utilization by 4.69%, 6.81%, and 8.27%, respectively, and improving load balancing by 33.36%, 19.69%, and 59.06%, respectively in average.

The most common neurological brain issue is Alzheimer’s disease, which can be diagnosed using a variety of clinical methods. However, the electroencephalogram (EEG) is shown to be effective in detecting Alzheimer’s disease. The purpose of this research is to develop a computer-aided diagnosis system that can diagnose Alzheimer’s disease using EEG data. In the present study, a band-pass elliptic digital filter was used to eliminate interference and disturbances from the EEG dataset. Next, the Discrete Wavelet Transform (DWT) technique has been employed to decompose the filtered signal into its frequency bands in order to extract the features of EEG signals. Then, different signal features such as logarithmic band power, standard deviation, variance, kurtosis, average energy, root mean square, and Norm have been integrated into the DWT technique to generate the feature vectors and improve the diagnosis performance. After that, nine machine learning approaches have been investigated to classify EEG features into their corresponding classes: linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), support vector machine (SVM), Naïve bayes (NB), k-nearest neighbor (KNN), decision tree (DT), extreme learning machine (ELM), artificial neural network (ANN), and random forests (RF). Finally, the performance of the different proposed machine learning approaches have been compared and evaluated by computing the sensitivity, specificity, overall diagnosis accuracy, and area under the receiver operating characteristic (ROC) curves and plotting the ROC curves and confusion matrices for five classification problems. These investigations aim to compare the proposed approaches and recommend the best combination method for the diagnosis of Alzheimer’s disorders. According to the results, the KNN classifier achieved an average classification accuracy of 99.98% with an area under the ROC curve of 100%. Our findings show that the suggested methodologies are an appealing supplementary tool for identifying possible biomarkers to help in the clinical diagnosis of Alzheimer’s disease.

This work presents an integrated framework for imputation of missing values and prediction of class label of unseen samples by using the best features of rule based inductive decision tree (DT) and Support Vector Machine (SVM) classifier (DT-SVM). In this work, the decision tree is used for imputation of missing values of the datasets containing both categorical and numerical valued attributes. In addition, some of the other popular and simple missing value imputation techniques like drop, mean, median, mode, and k-nearest neighbor (kNN) are used for a comparative analysis. The imputed datasets are then classified using SVM. The performance of the proposed integrated novel framework DT-SVM has been compared with Drop-SVM, Mean-SVM, Median-SVM, Mode-SVM, and kNN-SVM and it is found that DT-SVM outperforms others. Further, a new variant of kNN named it as approximated kNN (A-kNN) has been proposed to overcome some of the shortcomings of canonical kNN while learning from a training set imputed by DT. Unlike canonical kNN, A-kNN does not scan the entire training set. Instead, it processes some of the representative instances from the training dataset to identify the nearest neighbor. The class centroid approach is adopted to find the representative instances of the training set. The effectiveness in term of accuracy as well as computational time of A-kNN is examined by comparing with canonical kNN. It is found that computational time of the proposed A-kNN is drastically reduced as compared to canonical kNN without compromising with the classification accuracy.

The length of hospital stay and its implications have a significant economic and human impact. As a consequence, the prediction of that key parameter has been subject to previous research in recent years. Most previous work has analysed length of stay in particular hospital departments within specific study groups, which has resulted in successful prediction rates, but only occasionally reporting predictive patterns. In this work we report a predictive model for length of stay (LOS) together with a study of trends and patterns that support a better understanding on how LOS varies across different hospital departments and specialties. We also analyse in which hospital departments the prediction of LOS from patient data is more insightful. After estimating predictions rates, several patterns were found; those patterns allowed, for instance, to determine how to increase prediction accuracy in women admitted to the emergency room for enteritis problems. Overall, concerning these recognised patterns, the results are up to 21.61% better than the results with baseline machine learning algorithms in terms of error rate calculation, and up to 23.83% in terms of success rate in the number of predicted which is useful to guide the decision on where to focus attention in predicting LOS.

Human-Machine Interfaces employing biosignal-based inputs are hard to translate to real-life applications, in part because of the difficulty of developing generalized models to classify physiological events representing a user’s actions. In the proposed framework, an Electrooculography (EOG)-based game is operated through a pipeline of decision methods. These include a user-independent classification model of eye movements using a Convolutional Neural Network (CNN), which is fed with images created from signal windows, and an Ensemble of Utility Decision Networks (EUDN), which moderates the impact of oftentimes conflicting ocular events while enabling a more natural level of control over the interface. The CNN and the EUDN replace the normally used feature-based ocular event detection methods for EOG. Finally, a Reinforcement Learning-based game actuation approach simultaneously updates multiple $(State, Action)$ pairs for each rewarded outcome, intervenes to mitigate the consequences of wrongful game Commands, and can be used as part of a “shared-control” paradigm based on EOG. Results show a positive impact of Reinforcement Learning both in improving participants’ game performance as well as in reducing some of their subjective workload indicators.

Design defects affect project quality and hinder development and maintenance. Consequently, experts need to minimize these defects in software systems. A promising approach is to apply the concepts of refactoring at higher level of abstraction based on UML diagrams instead of code level. Unfortunately, we find in literature many defects that are described textually and there is no consensus on how to decide if a particular design violates model quality. Defects could be quantified as metrics based rules that represent a combination of software metrics. However, it is difficult to find manually the best threshold values for these metrics. In this paper, we propose a new approach to identify design defects at the model level using the ID3 decision tree algorithm. We aim to create a decision tree for each defect. We experimented our approach on four design defects: The Blob, Data class, Lazy class and Feature Envy defect, using 15 Object-Oriented metrics. The rules generated using decision tree give a very promising detection results for the four open source projects tested in this paper. In Lucene 1.4 project, we found that the precision is 67% for a recall of 100%. In general, the accuracy varies from 49%, reaching for Lucene 1.4 project 80%.

Diabetes is a major public health challenge affecting more than 451 million people. Physiological and experimental factors influence the accuracy of non-invasive glucose monitoring, and these need to be overcome before replacing the finger prick method. Also, the suitable employment of machine learning techniques can significantly improve the accuracy of glucose predictions. One aim of this study is to use light sources with multiple wavelengths to enhance the sensitivity and selectivity of glucose detection in an aqueous solution. Multiple wavelength measurements have the potential to compensate for errors associated with inter- and intra-individual differences in blood and tissue components. In this study, the transmission measurements of a custom built optical sensor are examined using 18 different wavelengths between 410 and 940 nm. Results show a high correlation value (0.98) between glucose concentration and transmission intensity for four wavelengths (485, 645, 860 and 940 nm). Five machine learning methods are investigated for glucose predictions. When regression methods are used, 9% of glucose predictions fall outside the correct range (normal, hypoglycemic or hyperglycemic). The prediction accuracy is improved by applying classification methods on sets of data arranged into 21 classes. Data within each class corresponds to a discrete 10 mg/dL glucose range. Classification based models outperform regression, and among them, the support vector machine is the most successful with F1-score of 99%. Additionally, Clarke error grid shows that 99.75% of glucose readings fall within the clinically acceptable zones. This is an important step towards critical diagnosis during an emergency patient situation.

Cardiovascular diseases (CVD) are among the most common serious illnesses affecting human health. CVDs may be prevented or mitigated by early diagnosis, and this may reduce mortality rates. Identifying risk factors using machine learning models is a promising approach. We would like to propose a model that incorporates different methods to achieve effective prediction of heart disease. For our proposed model to be successful, we have used efficient Data Collection, Data Pre-processing and Data Transformation methods to create accurate information for the training model. We have used a combined dataset (Cleveland, Long Beach VA, Switzerland, Hungarian and Stat log). Suitable features are selected by using the Relief, and Least Absolute Shrinkage and Selection Operator (LASSO) techniques. New hybrid classifiers like Decision Tree Bagging Method (DTBM), Random Forest Bagging Method (RFBM), K-Nearest Neighbors Bagging Method (KNNBM), AdaBoost Boosting Method (ABBM), and Gradient Boosting Boosting Method (GBBM) are developed by integrating the traditional classifiers with bagging and boosting methods, which are used in the training process. We have also instrumented some machine learning algorithms to calculate the Accuracy (ACC), Sensitivity (SEN), Error Rate, Precision (PRE) and F1 Score (F1) of our model, along with the Negative Predictive Value (NPR), False Positive Rate (FPR), and False Negative Rate (FNR). The results are shown separately to provide comparisons. Based on the result analysis, we can conclude that our proposed model produced the highest accuracy while using RFBM and Relief feature selection methods (99.05%).

Since the decision trees (DTs) have an advantage over “black-box” models, such as neural nets or support vector machines, in terms of comprehensibility, such that it might merit improvement for further optimization. The node splitting measures and pruning methods are primary among the techniques that can improve the generalization abilities of DTs. Here, we introduced the unequal interval optimization for node splitting, as well as the local chi-square test for tree pruning. This new method was named an adaptive multi-branch decision tree (CMDT). 11 benchmark data sets with different scales were chosen from UCI Machine Learning Repository and coupled with 12 classifiers to evaluate the CMDT algorithm. The results showed that CMDT can be more reliable than the twelve comparative approaches, especially for imbalanced datasets. We also discussed the performance metrics and the weighted decision-making table in unbalanced data sets. The CMDT algorithm can be found here: https://github.com/chenyuan0510/CMDT.

This paper presents the synthesis of multiband frequency selective surfaces (FSSs) using supervised machine learning (ML) with the decision tree (DT) algorithm. The proposed FSS structure is composed of an array of metallic patches printed on a dielectric substrate for stopband spatial filtering microwave applications. The shapes of the metallic patches are based on the sunflower (helianthus annus) geometry. In the first step, a parametric analysis is performed to investigate the use of different FSS geometries, including those with circular, annular and corolla integrated patch elements, to compose the sunflower geometry, regarding multiband and polarization independent performances with size reduction. Two bioinspired FSS geometries are synthesized using supervised machine learning with the decision tree algorithm. The random forest (RF) algorithm is used to validate the decision tree algorithm and to confirm the obtained results. The numerical analysis of the proposed FSS geometries is performed using Ansoft Designer software. Prototypes are fabricated and measured. The good agreement observed between simulated and measured results has validated the proposed approach. The use of supervised machine learning with the decision tree algorithm resulted in a particularly efficient and accurate synthesis procedure due to its intuitive implementation and simplified and effective data analysis modelling.

In this article, a machine learning aided electronic warfare (EW) system is presented and the simulation results are discussed. The developed EW system uses an automatic decision tree generator to create engagement protocol and a fuzzy logic model to quantify threat levels. A long-short term memory (LSTM) neural network was also trained to predict the next signal set of multifunction radars. The simulation results demonstrate the effectiveness of the developed EW system’s ability to engage multiple multifunction radars.

A fast Coding Unit (CU) partition decision strategy based on Human Visual System (HVS) perception quality is proposed in this paper. Considering that it is difficult for existing fast algorithms to further improve compression efficiency, perceptual coding technology has been tried to remove visual redundancy for achieving the purpose of reducing the bit rate on the basis of maintaining subjective visual quality. However, the existing perceptual coding model is still insufficient to reflect the characteristics of the HVS, which has limited improvement in coding efficiency, especially in fast algorithms. In this method, the characteristics of the limited human capacity for spatial-temporal resolution and the visual sensory memory are used to improve coding performance. First, the color complexity is used as a control factor to optimize the Just Noticeable Difference (JND) model to remove visual redundancy in a way that is more in line with the characteristics of visual perception. Second, a classification model of motion patterns based on human visual saliency is designed to provide a basis for CU classification, which effectively improves the coding accuracy. Finally, an offline Decision Tree (DT) classifier is designed based on the above model, and texture features are incorporated into the classifier as another key attribute to further reduce the computational complexity. The results of performance evaluation confirm that the proposed method achieves significantly improved coding performance compared with original Versatile Test Model (VTM). Compared with existing algorithms, our method not only improves coding efficiency, but also improves subjective visual quality.

The Domain Name System (DNS) is among the most ubiquitous and important protocols for network communication; however, security concerns regarding DNS have been on the rise and demand for encrypted traffic has followed suit. Using a publicly available dataset, this work compares 10 different machine learning classifiers using stratified 10-fold cross-validation. The classifiers are used to determine the most effective and efficient way of detecting malicious DNS over Hypertext Transfer Protocol Secure (HTTPS) traffic, dubbed DoH traffic. Model performance is evaluated on Non-DoH vs. DoH traffic, then tested on benign vs. malicious DoH traffic. Additionally, this paper seeks to build upon existing research by removing noise and introducing feature selection methods and feature explainability to produce a better model for real-world deployment. After eliminating five overfitting features, our findings indicate that light gradient boosting machine (LGBM) yielded the highest accuracy to training time ratio while approaching 0% error using 20 top features.

In the production process of large-scale machinery and complex industries, the key performance indicator (KPI) prediction is an essential part of project scheduling and cost estimation. The continuous enrichment of sensor types and functions brings us massive soft-sensing parameters for regression, but also brings severe challenges to algorithm learning. In this paper, an improved ensemble feature selection based on decision tree (EFS-DT) strategy for KPI prediction is developed. On the one hand, the ensemble of multi-criteria filtering results broadens the selector’s perspective without the time cost of superposition. On the other hand, credibility and similarity analysis are designed to eliminate the concerns of Dempster’s combination rule about conflict. After re-evaluating the variable scores, more high-quality variables can be selected to build a more accurate and robust KPI prediction model. Finally, a realistic shield tunnel case in China is used to evaluate the feasibility and effectiveness of proposed approach.

Fault identification in Photovoltaic (PV) array is a contemporary research topic motivated by the higher penetration levels of PV systems in recent electrical grids. Therefore, this work aims to define an optimal Machine learning (ML) structure of automatic detection and diagnosis algorithm for common PV array faults, namely, permanent (Arc Fault, Line-to-Line, Maximum Power Point Tracking unit failure, and Open-Circuit faults), and temporary (Shading) under a wide range of climate datasets, fault impedances, and shading scenarios. To achieve the best-fit ML structure, three distinct ML classifiers are compared, namely, Decision Tree (DT) based on different splitting criteria, K-Nearest Neighbors (KNN) based on the different metrics of distance and weighting functions, and Support Vector Machine (SVM) based on different Kernel functions and multi-classification approaches. Also, Bayesian Optimization is adopted to assign the optimal hyperparameters to the fault classifiers. To investigate the performance of classifiers reported, both simulation and experimental case studies are carried out and presented.

Skyline queries identify skyline points, the minimal set of data points that dominate all other data points in a large dataset. The main challenge with skyline queries is executing the skyline query in the shortest possible time. To address and solve skyline query performance issues, we propose a decision tree-based method known as the decision tree-based comparator (DC). This method minimizes unnecessary dominance tests (i.e., pairwise comparisons) by constructing a decision tree based on the dominance testing. DC uses dominance relations that can be obtained from the decision rules of the decision tree to determine incomparability between data points. DC can also be easily applied to improve the performance of various existing skyline query methods. After describing the theoretical background of DC and applying it to existing skyline queries, we present the results of various experiments showing that DC can improve skyline query performance by up to 23.15 times.

A semiconductor or photoelectric manufacturer faces a more competitive market with small quantities of many products. These products require hundreds of processes for production, thereby generating huge manufacturing data. With the help of the Internet of Things (IoT) technology, the manufacturer can collect manufacturing process data in a timely manner. Due to the massive quantities of manufacturing process data, it has become difficult for manufacturers to determine the causes of product defects, by which machine, and by what manufacturing process (or recipe) parameters. This research proposes a six-step data-driven solution to this problem. The chi-square test of independence, the Apriori algorithm, and the decision tree method identify the process that is generating the defective products and extract rules to identify the lot identification of product defects and their associated manufacturing process parameters. An empirical study was conducted at an optical thin-film filter (TFF) company in Taiwan. Based on the data of the optical TFF production lines, the coating process was identified as the source of the defective products, and the extracted rules were validated and implemented. The product defect rate decreased from 20% to 5%. Hence, the proposed data-driven solution was found to be capable of helping manufacturers enhance their product yield.

Automation of healthcare facilities represents a challenging task of streamlining a highly information-intensive sector. Modern healthcare processes produce large amounts of data that have great potential for health policymakers and data science researchers. However, a considerable portion of such data is not captured in electronic format and hidden inside the paperwork. A major source of missing data in healthcare is paper-based clinical pathways (CPs). CPs are healthcare plans that detail the interventions for the treatment of patients, and thus are the primary source for healthcare data. However, most CPs are used as paper-based documents and not fully automated. A key contribution towards the full automation of CPs is their proper computer modeling and encoding their data with international clinical terminologies. We present in this research an ontology-based CP automation model in which CP data are standardized with SNOMED CT, thus enabling machine learning algorithms to be applied to CP-based datasets. CPs automated under this model contribute significantly to reducing data missingness problems, enabling detailed statistical analyses on CP data, and improving the results of data analytics algorithms. Our experimental results on predicting the Length of Stay (LOS) of stroke patients using a dataset resulting from an e-clinical pathway demonstrate improved prediction results compared with LOS prediction using traditional EHR-based datasets. Fully automated CPs enrich medical datasets with more CP data and open new opportunities for machine learning algorithms to show their full potential in improving healthcare, reducing costs, and increasing patient satisfaction.

Music visualization can present music information through the visual way of graphic images, which is helpful to improve the accuracy and effectiveness of music information communication. In view of the shortcomings in the current music visualization field, this paper combines K-means clustering, fusion decision tree and other mathematical statistical methods on the basis of music graphic images to construct a music visualization model based on graphic images and mathematical statistics. First, the application principles of Schlieren imaging and laser Doppler imaging in the visualization of music graphic images are described. Secondly, on the basis of music graphic images, K-means clustering method is used to perform cluster analysis on music visualization information. Finally, through the fusion decision tree method, the classification of music visual information is studied. The actual case analysis and performance test results show the superiority of the music visualization method based on graphic images and mathematical statistics. This method can provide a scientific reference model and basis for the modern music industry to establish new visualization systems using graphics, images and mathematical statistics.

In order to improve the classification performance of a single classification model, Multiple Classifier Systems (MCS) are used. One of the most common techniques utilizing multiple decision trees is the random forest, where diversity between base classifiers is obtained by bagging the training dataset. In this paper, we propose the algorithm that uses horizontal partitioning the learning set and uses decision trees as base models to obtain decision regions. In the proposed approach feature space is divided into disjoint subspace. Additionally, the location of the subspace centroids, as well as the size and location of decision regions, are used in order to determine the weights needed in the last process of creating MCS, i.e. in the integration phase. The proposed algorithm was evaluated employing multiple open-source benchmarking datasets, compared using accuracy and Matthews correlation coefficient performance measures with two existing MCS methods - random forest and majority voting. The statistical analysis confirms an improvement in recognition compared to the random forest. In addition, we proved that for infinitely dense space division proposed algorithm is equivalent to majority voting.

In June 4, 2020, Corona Virus Disease 2019(COVID-19) cases in Wuhan were cleared, and the epidemic situation was basically controlled. Such public safety infectious disease includes influences great pressure on the national economy. At present, some countries and regions in the world are still in epidemic situation, and there is an urgent need to judge the infection situation and travel risk in the region. In a relatively fine scale down to perceive the surrounding situation, and then rational zoning decisions to promote the resumption of production and work. In this study, indicators for the evaluation of COVID-19 epidemic were constructed using multi-sourced data. A computational evaluation of 736 fine-grained grids was performed using the GeoDetector model and the decision tree model. The study found that the risk level in older neighborhoods was much higher than in newer neighborhoods; the population density was the most important determinant of infection; the number of urban people slumped to 37% of that in usual times according to Tencent data after the “city closure”; The model this paper used portrays the major factor in defining low-risk areas and high-risk areas, and offers suggestions and assessment from a geographical perspective to fight COVID-19, thus presenting great practical value.