Model Selection Research Articles

A Comparative Discussion on the Selection of Cardiac Hypertrophy Models: TAC Surgery Vs Ang II Infusion [Letter

A Comparative Discussion on the Selection of Cardiac Hypertrophy Models: TAC Surgery Vs Ang II Infusion [Letter

An exploratory study on automatic identification of assumptions in the development of deep learning frameworks

ContextStakeholders constantly make assumptions in the development of deep learning (DL) frameworks. These assumptions are related to various types of software artifacts (e.g., requirements, design decisions, and technical debt) and can turn out to be invalid, leading to system failures. Existing approaches and tools for assumption management usually depend on manual identification of assumptions. However, assumptions are scattered in various sources (e.g., code comments, commits, pull requests, and issues) of DL framework development, and manually identifying assumptions has high costs (e.g., time and resources). ObjectiveThe objective of the study is to evaluate different classification models for the purpose of identification with respect to assumptions from the point of view of developers and users in the context of DL framework projects (i.e., issues, pull requests, and commits) on GitHub. MethodFirst, we constructed a new and largest dataset (i.e., the AssuEval dataset) of assumptions collected from the TensorFlow and Keras repositories on GitHub. Then we explored the performance of seven non-transformers based models (e.g., Support Vector Machine, Classification and Regression Trees), the ALBERT model, and three decoder-only models (i.e., ChatGPT, Claude, and Gemini) for identifying assumptions on the AssuEval dataset. ResultsThe study results show that ALBERT achieves the best performance (f1-score: 0.9584) for identifying assumptions on the AssuEval dataset, which is much better than the other models (the 2nd best f1-score is 0.8858, achieved by the Claude 3.5 Sonnet model). Though ChatGPT, Claude, and Gemini are popular models, we do not recommend using them to identify assumptions in DL framework development because of their low performance. Fine-tuning ChatGPT, Claude, Gemini, or other language models (e.g., Llama3, Falcon, and BLOOM) specifically for assumptions might improve their performance for assumption identification. ConclusionsThis study provides researchers with the largest dataset of assumptions for further research (e.g., assumption classification, evaluation, and reasoning) and helps researchers and practitioners better understand assumptions and how to manage them in their projects (e.g., selection of classification models for identifying assumptions).

Machine learning discrimination and prediction of different quality grades of sauce-flavor baijiu based on biomarker and key flavor compounds screening

The quality grade of base Baijiu directly determines the final quality of sauce-flavor Baijiu. However, traditional methods for assessing these grades often rely on subjective experience, lacking objectivity and accuracy. This study used GC-FID, combined with quantitative descriptive analysis (QDA) and odor activity value (OAV), to identify 27 key flavor compounds, including acetic acid, propionic acid, ethyl oleate, and isoamyl alcohol etc., as crucial contributors to quality grade differences. Sixteen bacterial biomarkers, including Komagataeibacter and Acetobacter etc., and 7 fungal biomarkers, including Aspergillus and Monascus etc., were identified as key microorganisms influencing these differences. Additionally, reducing sugar content in Jiupei significantly impacted base Baijiu quality. Finally, 11 machine learning classification models and 9 prediction models were evaluated, leading to the selection of the optimal model for accurate quality grade classification and prediction. This study provides a foundation for improving the evaluation system of sauce-flavor Baijiu and ensuring consistent quality.

Residence time density (RTd) identification through deconvolution of finite time horizon input output data and its Application to a novel analytical dispersion process RTd model

This work presents first an approach to the identification of Residence Time density (RTd) and Residence Time Distribution (RTD) functions, through deconvolution of finite time horizon, input/output data generated by a zero-order hold input, using the novel concept of zero-order hold, average, impulse response coefficients. Subsequently, a novel RTd analytical expression is derived for an open-open axial dispersion process with a finite experimental section in a tube of infinite length, and an explanation is provided for this RTd model’s differences from the literature. This new RTd expression is then employed in two case studies to demonstrate the effectiveness of the proposed identification through a deconvolution conceptual framework in identifying RTd model parameters, and in assisting in the selection of RTd models which are compatible with experimental data.

Adaptive evolution and functional significance of the PPARGC1A gene across diverse animal species.

Codon-based analyses of the PPARGC1A gene across 38 vertebrate species were deployed to elucidate patterns of evolutionary change. Employing maximum likelihood assessments through MEGA, we scrutinized 447 codon positions addressing the entire coding region, excluding positions mired by gaps or missing data. Distinct codons manifested variance in selection pressures, particularly codons 4, 11, 66, and 123, which exhibited positive dN-dS values suggestive of positive selection. Codon 137 displayed the most pronounced dN-dS value, signifying intensified selective advantage. Meanwhile, codons 30 and 90 portrayed near-neutral scores, indicative of purifying selection. Complementary computational methods (IFEL, REL, FUBAR, and SLAC) confirmed positive selection at specific codon sites, with varying degrees of corroboration. The integration of mixed-effect modeling (MEME) identified episodic diversifying selection, pinpointing codons that underwent selection episodes in certain lineages. Refined codon model selection lent insight into substitution rates, revealing nuanced degrees of evolutionary conservation among different codons. Supporting these genetic insights, the phylogenetic analysis highlighted relationships among the PPARGC1A sequences and domain analysis confirmed conserved features across species, while protein-protein interaction networks suggested a complex web of functional interdependencies. These findings advance our understanding of the PPARGC1A gene's evolutionary trajectory and underscore the gene's potential adaptive significance within diverse vertebrate lineages.

Super-resolution localization and orientation estimation of multiple dipole sound sources: From a maximum likelihood framework to wind tunnel validation

This paper investigates the identification of multiple dipole sound sources using sound pressures measured from a microphone array. The problem is addressed in the maximum likelihood (ML) framework, where the locations, orientations, and powers of multiple dipole sound sources are unknown parameters to be estimated. By the consistency property of ML, the estimated parameters converge to their actual values, which implies an asymptotically perfect spatial resolution, if a sufficiently high signal-to-noise ratio can be achieved. In order to reduce the dimension of the optimization problem of ML, the contribution of each dipole source to the measured pressures is assumed to be a latent variable and the ML problem is equivalently solved via the expectation–maximization (EM) algorithm, which iteratively and sequentially updates each source contribution and the associated sound source parameters. The number of sound sources can also be determined by the model selection approaches which add a penalty of model dimension to the ML objective function. The proposed method is assessed via a laboratory experiment where the sound field is produced by dipole speakers and a wind tunnel experiment where airframe aerodynamic noise is generated at a high Reynolds number. Experimental results show that the proposed method outperforms existing approaches in the sense of higher spatial resolution, more accurate localization, and the capacity to identify the orientations of multiple dipole sound sources.

Asthma and allergy trajectories in children based on combined parental report and register data.

Trajectories of asthma and allergy in children are heterogeneous and commonly derived from parental report of disease or clinical records. This study combined parental-reported and register-based dispensed medication data to characterize childhood trajectories of co-existing asthma, allergic rhinitis, and eczema. From a Swedish population-based birth cohort (N = 5654), survey responses collected at the age of 1, 4.5, 8, and 12 years were linked to dispensed medication register data for the period of 2-13 years. Trajectories were identified with latent class analysis. Statistical metrics and clinical interpretability guided the model selection. Nine distinct trajectories were identified: three asthma-dominated (early-onset remitting [n = 189, 3.3%], late-onset [n = 117, 2.1%], and persistent [n = 149, 2.6%]), two eczema-dominated (persistent [n = 190, 3.4%] and remitting [n = 432, 7.6%]), one allergic rhinitis-dominated (late-onset [n = 259, 4.6%]), two multimorbidity (mid-childhood asthma and late-onset allergic rhinitis [n = 144, 2.5%], and persistent eczema and late-onset allergic rhinitis [n = 90, 1.6%]), and one low-disease burden trajectory (n = 4084, 72.2%). Differences were seen across the trajectories in the proportion of parental report of disease and dispensed medication as well as by class and quantity of medication dispensed. Combined parental-reported and dispensed medication data enriches characterization of longitudinal trajectories of asthma and allergy in children by merging subjective experience of disease with healthcare utilization. The identified trajectories were characterized by distinct disease development and prescription patterns suggesting clinically differential morbidity burden.

A least-squares framework for developing interval type-2 fuzzy semantics

The developing of IT2 fuzzy semantics is critical for computing with words (CWW), but exiting approaches lack flexibility and fail to adapt user’s diversified demands. The study proposes a least-squares framework for designing CWW encoders that construct interval type-2 fuzzy sets (IT2 FSs) to represent the semantic meanings of linguistic words. In the least-squares framework, an CWW encoder is characterized by two elements: an intra-uncertain semantic mapping and an inter-uncertain semantic family. The intra-uncertain semantic mapping transforms data intervals into type-1 fuzzy sets (T1 FSs), then an optimal IT2 FS is derived from the inter-uncertain semantic family using the least-squares method. Furthermore, several intra-uncertain semantic mappings are introduced, and a compatibility measure is defined to facilitate model selection. The least-squares framework benefits from the flexible selection of intra-uncertain semantic mappings and least-squares optimization-based construction of IT2 FSs. In experiments, the least-squares framework is applied to handle real-world online survey data and the large-scale online review data set of a Chinese life service review site, Dianping.com. Compared to the enhanced interval approach and the Hao-Mendel approach, the least-squares framework shows its favorable efficiency in experiments and statistical tests, and adapts to user-defined intra- and inter-uncertain semantic families.

New Bounded Probability Model: Properties, Estimation, and Applications

The article introduces a new unit distribution, an extension of Tiessier distribution, characterized by a versatile hazard function capable of adopting diverse shapes such as bathtub or N-shaped curves. An exploration of the fundamental properties of this distribution is undertaken, accompanied by the implementation of the maximum likelihood estimation technique and eleven alternative methods to approximate its parameters effectively. Through a simulation study, the article effectively demonstrates the precision of these parameter estimation methods, even when dealing with small sample sizes. Two datasets are employed to apply the novel distribution, subjecting it to a comprehensive evaluation against established models utilizing a range of model selection criteria and goodness of fit tests. Notably, the article illustrates that the performance of the new distribution surpasses that of existing models in effectively capturing data patterns. Beyond its empirical contributions, the article highlights the potential cross-disciplinary applications of the new distribution in many fields while concurrently advancing the realms of probability theory and statistical inferences.

THE DEVELOPMENT OF LEARNING KITS BASED ON CAUSALITY MODEL TO IMPROVE CREATIVITY THINKING SKILLS MOMENTUM AND IMPULSE OF STUDENTS

Teachers still do not fully pay attention to the creative thinking abilities of students. This is seen from the selection of models in the teaching tools used by teachers in teaching. The selection of teaching models that tend to be directed towards direct teaching results in students being less active in following physics learning. Therefore, this research aims to determine the feasibility of a causality model learning device to improve creative thinking skills on the material of momentum and impulse. This type of research includes research and development (R&D), using a 4D model design. However, in this research, only the development stage is carried out. The research procedures carried out refer to the steps of developing the 4D model, namely define, design, and develop. The developed learning device consists of a syllabus, lesson plans, and pre-tasks. Data collection is done through validation of the learning device by validators. The results of the validation of the learning device reached an average validation value of 87.50% with the criteria of very valid and the reliability score is 92.30%. Based on these results, it can be concluded that the causality model learning device to improve the creative thinking skills of momentum and impulse students is suitable to be used in learning.

Identifying Brain Network Structure for an fMRI Effective Connectivity Study Using the Least Absolute Shrinkage and Selection Operator (LASSO) Method

Background: Studying causality relationships between different brain regions using the fMRI method has attracted great attention. To investigate causality relationships between different brain regions, we need to identify both the brain network structure and the influence magnitude. Most current methods concentrate on magnitude estimation, but not on identifying the connection or structure of the network. To address this problem, we proposed a nonlinear system identification method, in which a polynomial kernel was adopted to approximate the relation between the system inputs and outputs. However, this method has an overfitting problem for modelling the input–output relation if we apply the method to model the brain network directly. Methods: To overcome this limitation, this study applied the least absolute shrinkage and selection operator (LASSO) model selection method to identify both brain region networks and the connection strength (system coefficients). From these coefficients, the causality influence is derived from the identified structure. The method was verified based on the human visual cortex with phase-encoded designs. The functional data were pre-processed with motion correction. The visual cortex brain regions were defined based on a retinotopic mapping method. An eight-connection visual system network was adopted to validate the method. The proposed method was able to identify both the connected visual networks and associated coefficients from the LASSO model selection. Results: The result showed that this method can be applied to identify both network structures and associated causalities between different brain regions. Conclusions: System identification with LASSO model selection algorithm is a powerful approach for fMRI effective connectivity study.

Holistic learning and optimization scheme for wireless networks

AbstractFuture networks are expected to use Artificial Intelligence and machine learning applications intensely. The primary focus of machine learning techniques will be on intelligent automation and decision‐making based on information learned from network functions, resources, and users. The research community has studied several schemes for using machine learning in communications and networks in the last two decades. However, the schemes mainly focus on learning the state information of individual network functions and resources and performing their optimization in a standalone fashion. Recently, a study group from the International Telecommunication Union has outlined machine learning pipeline requirements. It addresses several aspects of machine learning operations, such as model selection, data sources, and action points. However, the mechanisms to define the state of network functions and resources for different network layers and domains are yet to be explored. Therefore, in addition to the above, specific studies are required for cross‐layer and administrative domain knowledge sharing and holistic optimization. Without considering the holistic view, applying machine learning to optimizing individual functions and resources may result in nonoptimal and unexpected behaviors instead of having any positive effect. The need for global and deep holistic learning has recently been identified in the literature; however, no implementation of the above scheme has been studied or evaluated for networks. This article proposes a novel approach, implementing global and deep holistic learning in wireless networks toward fulfilling the requirements of future intelligent networks. It also proposes an objective function‐based feature engineering for wireless networks. Experimental results of the proposed scheme show significant improvements in the network performance and accuracy of the machine learning models. Moreover, applications of transfer learning to base learners for network functions and resources can significantly expedite decision‐making with reduced computational costs.

Evaluasi Model Deep Learning pada Pola Dataset Biomedis

This study aims to evaluate the effectiveness and efficiency of various deep learning models in recognizing patterns within diverse biomedical datasets. The methods involved the collection of biomedical data from various public and synthetic sources, including chest radiographs, MRI, CT scans, as well as electrocardiogram (ECG) and electromyography (EMG) signals. The data underwent preprocessing steps such as normalization, noise removal, and data augmentation to improve quality and variability. The deep learning models evaluated included Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN), which were trained to identify patterns within the data. The performance evaluation was conducted using metrics like accuracy, sensitivity, specificity, and AUC to ensure the models' generalization capabilities on test datasets. The results revealed that CNNs excelled in medical image analysis, particularly in terms of accuracy and interpretability, while RNNs were more effective in handling sequential data such as medical signals. The primary conclusion of this study is that the selection of deep learning models should be tailored to the type of data and specific application requirements, emphasizing the importance of improving model interpretability and generalization for broader applications in clinical settings.

Determination Of Return On Assets: Analysis Of Solvency Ratio And Liquidity Ratio

The main objective of this study is to provide an in-depth understanding of how solvency and liquidity ratios affect a company's ability to use its assets efficiently in generating profits, as well as to make useful contributions to managerial decision making and evaluation of company performance. Several indicators in the study were studied to analyze the determination of the solvency ratio consisting of (Debt to Asset Ratio and Debt to Equity Ratio) and liquidity ratio (Current Ratio) to Return on Assets (ROA) with a sample of several large industrial companies listed on the Indonesia Stock Exchange with a period of 2015 2020. Research with test analysis techniques analysis of test selection of Test models (chow, hausman and langrange multiplier), classical assumption test with Multicollinearity and Heteroscedasticity, Regression Test panel data, Correlation Test and hypothesis testing (partial / t-Test and simultaneous / F-Test). The results of research through partial tests and simultaneous tests with the main focus of industrial companies resulted in that the three variations of DAR, DER and CR had no influence on the Return on Assets of the company objects sampled.

The Dynamics Stability Judging Method for a Gantry-type Welding Robot

In order to improve the dynamic performance of gantry-type welding robot, this paper introduces the structural dynamics model and Lyapunov stability criterion of the gantry-type welding robot frame with nine degrees of freedom. The dynamic equations of the welding robot satisfy the Euler–Lagrange form, and the calculation index of welding robot Lyapunov stability is judged by matrix eigenvalues. Combining with the specific design parameters of an industrial welding robot, the motion stability of the gantry-type welding robot frame and the displacement response of the welding torch are analyzed. The influence of the change of the frame design structure on the stability of the welding torch for end effector of welding robot is discussed, furthermore, the improvement direction of vibration damping and anti-interference of welding robot system is pointed out. Finally, the dynamics simulation of the optimized gantry-type welding robot end effector is carried out by using ADAMS software, and the torque variation curve of the end effector is obtained, which provides a strong basis for the selection of motor models and the design of control system.

Reinforcement learning in autonomous defense systems: Strategic applications and challenges

Reinforcement learning (RL) is one of the most progressive ways to improve the efficiency of drones and robotic systems in the context of the defense and security industries. In Anomaly Detection (AD) systems, this paper discusses the potential of RL as a Model Selection Criteria (MSC) method and its associated issues. RL algorithms enable such systems to adapt to and learn from interactions with their environment, optimizing their response to complex and dynamic threats. This way, new states may be introduced with better decision-making processes, improving the effectiveness of the operations and enabling Reinforcement learning -powered systems to learn new scenarios as well as perform detailed defensive actions in a real-time context. The importance of RL in national defense can therefore be said to lie in its capacities for changing how threats are identified, how threats are responded to, and even what strategies of defense are thought of. Self-governing systems that incorporate Reinforcement learning are capable of functioning in an unpredictable environment, evaluating threats correctly, and carrying out defensive measures with the help of people practically at all. This versatility is critical in modern warfare because a preferred option is the right response to threats that were unknown a few years ago. However, incorporating RL into autonomous defense systems is not without big challenges. Some important problem areas are improved stability and accuracy of RL algorithms in various and critical situations, the legal and practical consequences of autonomous decision-making and possible threats that can arise due to adversarial manipulation of learning algorithms. In addition, such systems have to be developed and put into practice under national and international standards to meet certain requirements and build confidence in such applications. This paper explores these strategic uses and issues and presents a detailed overview of how RL can improve independent defense functionalities, as well as the significant problems relating to this approach. Thus, by providing case studies and theoretical analysis, it aims to demonstrate that RL has the potential to define the technological development of defense and benefit national security systems in the context of a growing threat.

Landslide Susceptibility Assessment Based on Multisource Remote Sensing Considering Inventory Quality and Modeling

The completeness of landslide inventories and the selection of evaluation models significantly impact the accuracy of landslide susceptibility assessments. Conventional field geological survey methods and single remote-sensing technology struggle to reliably identify landslides under complex environmental conditions. Moreover, prevalent landslide susceptibility evaluation models are often plagued by issues such as subjectivity and overfitting. Therefore, we investigated the uncertainty in susceptibility modeling from the aspects of landslide inventory quality and model selection. The study focused on Luquan County in Yunnan Province, China. Leveraging multisource remote-sensing technologies, particularly emphasizing optical remote sensing and InSAR time-series deformation detection, the existing historical landslide inventory was refined and updated. This updated inventory was subsequently used to serve as samples. Nine evaluation indicators, encompassing factors such as distance to faults and tributaries, lithology, distance to roads, elevation, slope, terrain undulation, distance to the main streams, and average annual precipitation, were selected on the basis of the collation and organization of regional geological data. The information value and two coupled machine-learning models were formulated to evaluate landslide susceptibility. The evaluation results indicate that the two coupled models are more appropriate for susceptibility modeling than the single information value (IV) model, with the random forest model optimized by genetic algorithm in Group I2 exhibiting higher predictive accuracy (AUC = 0.796). Furthermore, comparative evaluation results reveal that, under equivalent model conditions, the incorporation of a remote-sensing landslide inventory significantly enhances the accuracy of landslide susceptibility assessment results. This study not only investigates the impact of landslide inventories and models on susceptibility outcomes but also validates the feasibility and scientific validity of employing multisource remote-sensing technologies in landslide susceptibility assessment.

Redefining Spectral Data Analysis with Immersive Analytics: Exploring Domain-Shifted Model Spaces for Optimal Model Selection.

Modern developments in autonomous chemometric machine learning technology strive to relinquish the need for human intervention. However, such algorithms developed and used in chemometric multivariate calibration and classification applications exclude crucial expert insight when difficult and safety-critical analysis situations arise, e.g., spectral-based medical decisions such as noninvasively determining if a biopsy is cancerous. The prediction accuracy and interpolation capabilities of autonomous methods for new samples depend on the quality and scope of their training (calibration) data. Specifically, analysis patterns within target data not captured by the training data will produce undesirable outcomes. Alternatively, using an immersive analytic approach allows insertion of human expert judgment at key machine learning algorithm junctures forming a sensemaking process performed in cooperation with a computer. The capacity of immersive virtual reality (IVR) environments to render human comprehensible three-dimensional space simulating real-world encounters, suggests its suitability as a hybrid immersive human-computer interface for data analysis tasks. Using IVR maximizes human senses to capitalize on our instinctual perception of the physical environment, thereby leveraging our innate ability to recognize patterns and visualize thresholds crucial to reducing erroneous outcomes. In this first use of IVR as an immersive analytic tool for spectral data, we examine an integrated IVR real-time model selection algorithm for a recent model updating method that adapts a model from the original calibration domain to predict samples from shifted target domains. Using near-infrared data, analyte prediction errors from IVR-selected models are reduced compared to errors using an established autonomous model selection approach. Results demonstrate the viability of IVR as a human data analysis interface for spectral data analysis including classification problems.

Efficient approximation of global population dynamic models through statistical inference using local data

Biological growth curves are pivotal in predicting natural growth across disciplines, typically analyzed using nonlinear least squares or maximum likelihood methods. Bhowmick et al. (2014) introduced the interval-specific rate of parameters (ISRP) for growth equations, improving the estimation of relative growth rate (RGR) and model selection accuracy. Despite its effectiveness, computing these model-specific RGR estimates involves complex calculations and lacks explicit expressions for many nonlinear models. Also, for highly nonlinear models and non-monotonic data where the parameters are non-linearly related, the computation of interval estimates is almost impossible and may suffer from significant approximation errors. So, the need for a more efficient computation method for ISRP remains a significant challenge in growth studies. In this article, we propose a computational approach to obtain interval estimates of parameters based on the maximum likelihood estimation method. The likelihood function is maximized using the data on smaller intervals. Our study underscores the importance of an efficient ISRP computation technique, providing a more stable, unbiased, and normally distributed estimator. The most important advantage is that it can be implemented using existing optimizers in software packages efficiently, therefore, giving more accessibility to the practitioners. Both simulation studies and real data analysis have been carried out to validate the proposed estimation process. Additionally, its applicability to non-monotonic growth profiles and its robustness in handling highly non-linear growth equations highlight its versatility. We also developed a web application GpEM-R which is freely available for researchers and practitioners to analyze growth data.

Classification of mindfulness experiences from gamma-band effective connectivity: Application of machine-learning algorithms on resting, breathing, and body scan

Background and ObjectivePracticing mindfulness is a mental process toward interoceptive awareness, achieving stress reduction and emotion regulation through brain-function alteration. Literature has shown that electroencephalography (EEG)-derived connectivity possesses the potential to differentiate brain functions between mindfulness naïve and mindfulness experienced, where such quantitative differentiation could benefit telediagnosis for mental health. However, there is no prior guidance in model selection targeting on the mindfulness-experience prediction. Here we hypothesized that the EEG effective connectivity could reach a good prediction performance in mindfulness experiences with brain interpretability. MethodsWe aimed at probing direct Directed Transfer Function (dDTF) to classify the participants’ history of mindfulness-based stress reduction (MBSR), and aimed at optimizing the prediction accuracy by comparing multiple machine learning (ML) algorithms. Targeting the gamma-band effective connectivity, we evaluated the EEG-based prediction of the mindfulness experiences across 7 machine learning (ML) algorithms and 3 sessions (i.e., resting, focus-breathing, and body-scan). ResultsThe support vector machine and naïve Bayes classifiers exhibited significant accuracies above the chance level across all three sessions, and the decision tree algorithm reached the highest prediction accuracy of 91.7 % with the resting state, compared to the classification accuracies with the other two mindful states. We further conducted the analysis on essential EEG channels to preserve the classification accuracy, revealing that preserving just four channels (F7, F8, T7, and P7) out of 19 yielded the accuracy of 83.3 %. Delving into the contribution of connectivity features, specific connectivity features predominantly located in the frontal lobe contributed more to classifier construction, which aligned well with the existing mindfulness literature. ConclusionIn the present study, we initiated a milestone of developing an EEG-based classifier to detect a person's mindfulness experience objectively. The prediction accuracy of the decision tree was optimal to differentiate the mindfulness experiences using the local resting-state EEG data. The suggested algorithm and key channels on the mindfulness-experience prediction may provide guidance for predicting mindfulness experiences using the EEG-based classification embedded in future wearable neurofeedback systems or plausible digital therapeutics.