Learning Classifiers Research Articles

EEG based functional connectivity in resting and emotional states may identify major depressive disorder using machine learning

ObjectiveDisrupted brain network connectivity underlies major depressive disorder (MDD). Altered EEG based Functional connectivity (FC) with Emotional stimuli in major depressive disorder (MDD) in addition to resting state FC may help in improving the diagnostic accuracy of machine learning classification models. We explored the potential of EEG-based FC during resting state and emotional processing, for diagnosing MDD using machine learning approach. MethodsEEG was recorded during resting state and while watching emotionally contagious happy and sad videos in 24 drug-naïve MDD patients and 25 healthy controls. FC was quantified using the Phase Lag Index. Three Random Forest classifier models were constructed to classify MDD patients and healthy controls, Model-I incorporating FC features from the resting state and Model-II and Model-III incorporating FC features while watching happy and sad videos respectively. ResultsImportant features distinguishing MDD and healthy controls were from all frequency bands and represent functional connectivity between fronto-temporal, fronto-parietal and fronto occipital regions. The cross-validation accuracies for Model-I, Model-II and Model-III were 92.3%, 94.9% and 89.7% and test accuracies were 60%, 80% and 70% respectively. Incorporating emotionally contagious videos improved the classification accuracies. ConclusionFindings support EEG FC patterns during resting state and emotional processing along with machine learning can be used to diagnose MDD. Future research should focus on replicating and validating these results. SignificanceEEG FC pattern combined with machine learning may be used for assisting in diagnosing MDD.

Research of 2D-COS with metabolomics modifications through deep learning for traceability of wine

To tackle the difficulty of extracting features from one-dimensional spectral signals using traditional spectral analysis, a metabolomics analysis method is proposed to locate two-dimensional correlated spectral feature bands and combine it with deep learning classification for wine origin traceability. Metabolomics analysis was performed on 180 wine samples from 6 different wine regions using UPLC-Q-TOF-MS. Indole, Sulfacetamide, and caffeine were selected as the main differential components. By analyzing the molecular structure of these components and referring to the main functional groups on the infrared spectrum, characteristic band regions with wavelengths in the range of 1000–1400 nm and 1500–1800 nm were selected. Draw two-dimensional correlation spectra (2D-COS) separately, generate synchronous correlation spectra and asynchronous correlation spectra, establish convolutional neural network (CNN) classification models, and achieve the purpose of wine origin traceability. The experimental results demonstrate that combining two segments of two-dimensional characteristic spectra determined by metabolomics screening with convolutional neural networks yields optimal classification results. This validates the effectiveness of using metabolomics screening to determine spectral feature regions in tracing wine origin. This approach effectively removes irrelevant variables while retaining crucial chemical information, enhancing spectral resolution. This integrated approach strengthens the classification model's understanding of samples, significantly increasing accuracy.

A data-driven and cost-oriented FMEA-MCDM approach to risk assessment and ranking in a fuzzy environment: A hydraulic pump factory case study.

In today's highly competitive business environment, firms strive to maximize profitability by minimizing or eliminating disruptions and failures to maintain a competitive edge. This study focuses on evaluating risks in a hydraulic pump factory as a means to achieve sustainable growth. To accomplish this, a team of experts was formed to identify potential errors, utilizing a combination of risk priority number criteria weighted by Fuzzy Shannon's entropy and a fusion of multi-criteria decision-making and failure mode and effects analysis for evaluating and ranking failures. Moreover, the study emphasizes the importance of considering the interaction among risk assessment indicators, the inclusion of cost of failure, and modeling under fuzzy uncertainty circumstances, as they have a notable impact on the final ranking of failures to be processed for risk mitigation action planning. This research brings a new dimension to enhance the overall effectiveness of risk assessment by aggregation, as evidenced by a novel use of data classification in machine learning and correlation in statistics. The findings indicate that the aggregated ranking data series is best matched and most influenced by the weighted aggregated sum product assessment method, with the highest rate of recall and precision accomplished.

A Vision Transformer-Based Framework for Knowledge Transfer From Multi-Modal to Mono-Modal Lymphoma Subtyping Models.

Determining lymphoma subtypes is a crucial step for better patient treatment targeting to potentially increase their survival chances. In this context, the existing gold standard diagnosis method, which relies on gene expression technology, is highly expensive and time-consuming, making it less accessibility. Although alternative diagnosis methods based on IHC (immunohistochemistry) technologies exist (recommended by the WHO), they still suffer from similar limitations and are less accurate. Whole Slide Image (WSI) analysis using deep learning models has shown promising potential for cancer diagnosis, that could offer cost-effective and faster alternatives to existing methods. In this work, we propose a vision transformer-based framework for distinguishing DLBCL (Diffuse Large B-Cell Lymphoma) cancer subtypes from high-resolution WSIs. To this end, we introduce a multi-modal architecture to train a classifier model from various WSI modalities. We then leverage this model through a knowledge distillation process to efficiently guide the learning of a mono-modal classifier. Our experimental study conducted on a lymphoma dataset of 157 patients shows the promising performance of our mono-modal classification model, outperforming six recent state-of-the-art methods. In addition, the power-law curve, estimated on our experimental data, suggests that with more training data from a reasonable number of additional patients, our model could achieve competitive diagnosis accuracy with IHC technologies. Furthermore, the efficiency of our framework is confirmed through an additional experimental study on an external breast cancer dataset (BCI dataset).

Teaching classical machine learning as a graduate-level course in chemical engineering: An algorithmic approach

The demand for engineering graduates with technical skills in data science, machine learning (ML), and artificial intelligence (AI) is now growing. Chemical engineering (ChemE) departments around the world are currently addressing this skills gap by instituting AI or ML elective courses in their program. However, designing such a course is difficult since the issue of which ML models to teach and the depth of theory to be discussed remains unclear. In this paper, we present a graduate-level ML course particularly designed such that students will be able to apply ML for research in ChemE. To achieve this, the course intends to cover a wide selection of ML models with emphasis on their motivations, derivations, and training algorithms, followed by their applications to ChemE-related data sets. We argue that this algorithmic approach to teaching ML can help broaden the capabilities of students since they can judge for themselves which tool to use when, even for problems outside the process industries, or they can modify the methods to test novel ideas. We found that students remain engaged in the mathematical details as long as every topic is properly motivated and the gaps in the required statistical and computer science concepts are filled. Hence, this paper also presents a roadmap of ML topics, their motivations, and bridging topics that can be followed by instructors. Lastly, we report anonymized student feedback on this course which is being offered at the Department of Chemical Engineering, University of the Philippines, Diliman.

Prediction of failures in sewer networks using various machine learning classifiers

ABSTRACT In sewer networks, failure prediction plays a significant role in operation and maintenance plans of wastewater utilities. This study aims to determine the effective variables on the failures by using feature selection algorithms (FS) and achieve maximum model accuracy with minimum variables. Also, four scenarios based on the suggested FS algorithms were developed. In these scenarios, the best prediction models were investigated using machine learning classifiers (ML) such as neural network classifier (NNC), gradient boosting machine (GBM), random forest (FR), and hybrid model (HM). The classification performance of ML models was evaluated using accuracy, precision, F1_score, and receiver operating characteristics (ROC) curve. The model accuracies ranging from 0.99 for accuracy to 1 for the ROC curve were achieved through ML algorithms. In conclusion, the ML algorithms suggested in this study may be a decision support tool for wastewater utilities in prioritizing the replacement, maintenance, and inspection of sewer pipes.

Medicinal Herb Identification Using CNN Model

Abstract: Automated herb identification plays a crucial role in various industries such as cosmetics, medicine, and food, where the need to accurately identify different plant species is essential. However, existing methods often face challenges when dealing with complex backgrounds and a wide variety of patterns, especially in wild environments. In response to these challenges, we propose an innovative convolutional neural network (CNN) model that incorporates two key components: the Part- Information Perception Module and the Species Classification Module. The Part-Information Perception Module in our model is designed to focus the model's attention on the relevant parts of the herb, effectively suppressing background noise. This mechanism allows the model to better discern the distinctive features of the herb itself, improving overall accuracy. Furthermore, we employ depthwise separable convolution and label smoothing techniques to reduce model complexity and minimize the impact of labelling inconsistencies, enhancing both efficiency and reliability. To validate the effectiveness of our model, we conducted experiments using a diverse dataset of herb images. The results demonstrate a significant improvement in accuracy and model efficiency compared to existing methods, making it a valuable tool for herb identification in challenging environments. In addition to herb identification, we also address the complex task of recognizing medicinal plant species. To facilitate this, we have created a dataset comprising image features of plant leaves, enabling the development of an automated recognition system using machine learning classifiers. Our experiments show that the classifiers achieve an impressive average accuracy rate of over 97%, highlighting the potential of our approach in accurately identifying medicinal plant species.

Implementation of Ensemble Machine Learning Classifier and Synthetic Minority Oversampling Technique for Sentiment Analysis of Sustainable Development Goals in Indonesia

As part of the Sustainable Development Goals (SDGs), governments worldwide have committed to improving people's lives to improve the quality of life for all, including the 17 such goals that were agreed upon in 2015 to benefit the human race as a whole. It would be interesting to see how society responds to the SDGs after approximately half of them have been achieved. This public response was analyzed in terms of sentiment. Within the total number of internet users in Indonesia, there are 18.45 million Twitter users. The platform enables anyone to write about anything they are experiencing in their lives, such as what is happening in their environment, what is happening in their education system, what is happening in the food industry, how people feel, and many more. The platform enables anyone to write about anything they are experiencing in their lives, such as what is happening in their environment, what is happening in their education system, what is happening in the food industry, how people feel, and many more. To model the data collected, the researchers used Ensemble Machine Learning Classifiers (EMLC) to model the data by using a machine learning classifier that uses machine learning techniques. The best model in this study is EMLC-Stacking with a data splitting of 80:20 and using SMOTE, which obtains an accuracy of 91%. This accuracy results from a 5% increase compared to when not using SMOTE. From 15,698 tweets, this research found that 47% were positive sentiments, 28% were negative sentiments, and 25% were neutral sentiments. The results that we measured offer hope that there will be a positive trend in the journey of the SDGs until 2030 if these findings are true.

Secure Your Network: Building a Hybrid Intrusion Detection System in Java

Abstract: To protect the system from different types of intrusions, an intrusion detection system ID is required. It is essential to analyse the communication in order to classify the material as malicious or helpful. The usage of intrusion detection systems for cyber security shouldn't add to the processing time required for classification. These days, classification algorithms are utilized in conjunction with machine learning approaches to identify harmful data or intrusions. KDD cup 99 is the data set that was used in the experiment. Hybrid classification models can be used to adjust the performance of individual classification methods. This model blends rule-based algorithms with categorization techniques. An additional degree of security is provided by the combination of machine and human intelligence in classification. Precision, recall, F-Measure, and Mean Age Precision are used to validate an algorithm. The algorithm has a 92.35 percent accuracy rate. Even after merging our human-written criteria with traditional machine learning classification techniques, the model's accuracy is still considered satisfactory. However, there is still room for improvement and a more specific classification of the attack

Deep-GenMut: Automated genetic mutation classification in oncology: A deep learning comparative study

Early cancer detection and treatment depend on the discovery of specific genes that cause cancer. The classification of genetic mutations was initially done manually. However, this process relies on pathologists and can be a time-consuming task. Therefore, to improve the precision of clinical interpretation, researchers have developed computational algorithms that leverage next-generation sequencing technologies for automated mutation analysis. This paper utilized four deep learning classification models with training collections of biomedical texts. These models comprise bidirectional encoder representations from transformers for Biomedical text mining (BioBERT), a specialized language model implemented for biological contexts. Impressive results in multiple tasks, including text classification, language inference, and question answering, can be obtained by simply adding an extra layer to the BioBERT model. Moreover, bidirectional encoder representations from transformers (BERT), long short-term memory (LSTM), and bidirectional LSTM (BiLSTM) have been leveraged to produce very good results in categorizing genetic mutations based on textual evidence. The dataset used in the work was created by Memorial Sloan Kettering Cancer Center (MSKCC), which contains several mutations. Furthermore, this dataset poses a major classification challenge in the Kaggle research prediction competitions. In carrying out the work, three challenges were identified: enormous text length, biased representation of the data, and repeated data instances. Based on the commonly used evaluation metrics, the experimental results show that the BioBERT model outperforms other models with an F1 score of 0.87 and 0.850 MCC, which can be considered as improved performance compared to similar results in the literature that have an F1 score of 0.70 achieved with the BERT model.

Data & Semantic Analysis Fake News Detection Based Domain-Specific Auto Encoder Learning

Abstract: Efforts to create technology for automatically detecting fake news are actively underway as the spread of misinformation on social media continues to escalate. However, most of these approaches primarily concentrate on the linguistic and structural aspects of fake news, such as identifying sources or authors, message length, and the frequency of negative language. In contrast, our study introduces a phony news detection model leveraging machine learning that incorporates user behaviors, news, and social capital-based social network dynamics. We used the XG Boost model to evaluate each feature's significance and identify the critical elements affecting the identification of fake news to capture the variables related to the spread of phony news entirely. We implemented SVM, RF, LR, CART, and NNET, well-known machine learning classification models, using these identified factors and assessed how well they detected bogus news. This work used a cross-validation process to prevent overfitting and to generalize the established models. Additionally, the predicted accuracy of the established models was compared. The RF model had the best forecast accuracy, about 94%, while the NNET had the worst performance rate, about 92.1%. As disinformation is created and disseminated with growing sophistication, the results of this study should improve the effectiveness of false news detection systems.

Return to Tradition: Learning Reliable Heuristics with Classical Machine Learning

Return to Tradition: Learning Reliable Heuristics with Classical Machine Learning

ER-GET: Emotion Recognition Based on Global ECG Trajectory.

In recent years, the recognition of human emotions based on electrocardiogram (ECG) signals has been considered a novel area of study among researchers. Despite the challenge of extracting latent emotion information from ECG signals, existing methods are able to recognize emotions by calculating the heart rate variability (HRV) features. However, such local features have drawbacks, as they do not provide a comprehensive description of ECG signals, leading to suboptimal recognition performance. For the first time, we propose a new strategy to extract hidden emotional information from the global ECG trajectory for emotion recognition. Specifically, a period of ECG signals is decomposed into sub-signals of different frequency bands through ensemble empirical mode decomposition (EEMD), and a series of multi-sequence trajectory graphs is constructed by orthogonally combining these sub-signals to extract latent emotional information. Additionally, to better utilize these graph features, a network has been designed that includes self-supervised graph representation learning and ensemble learning for classification. This approach surpasses recent notable works, achieving outstanding results, with an accuracy of 95.08% in arousal and 95.90% in valence detection. Additionally, this global feature is compared and discussed in relation to HRV features, with the intention of providing inspiration for subsequent research.

Multiclass skin lesion classification using deep learning networks optimal information fusion

A serious, all-encompassing, and deadly cancer that affects every part of the body is skin cancer. The most prevalent causes of skin lesions are UV radiation, which can damage human skin, and moles. If skin cancer is discovered early, it may be adequately treated. In order to diagnose skin lesions with less effort, dermatologists are increasingly turning to machine learning (ML) techniques and computer-aided diagnostic (CAD) systems. This paper proposes a computerized method for multiclass lesion classification using a fusion of optimal deep-learning model features. The dataset used in this work, ISIC2018, is imbalanced; therefore, augmentation is performed based on a few mathematical operations. After that, two pre-trained deep learning models (DarkNet-19 and MobileNet-V2) have been fine-tuned and trained on the selected dataset. After training, features are extracted from the average pool layer and optimized using a hybrid firefly optimization technique. The selected features are fused in two ways: (i) original serial approach and (ii) proposed threshold approach. Machine learning classifiers are used to classify the chosen features at the end. Using the ISIC2018 dataset, the experimental procedure produced an accuracy of 89.0%. Whereas, 87.34, 87.57, and 87.45 are sensitivity, precision, and F1 score respectively. At the end, comparison is also conducted with recent techniques, and it shows the proposed method shows improved accuracy along with other performance measures.

Unlocking the adaptive advantage: correlation and machine learning classification to identify optimal online adaptive stereotactic partial breast candidates

Objective. Online adaptive radiotherapy (OART) is a promising technique for delivering stereotactic accelerated partial breast irradiation (APBI), as lumpectomy cavities vary in location and size between simulation and treatment. However, OART is resource-intensive, increasing planning and treatment times and decreasing machine throughput compared to the standard of care (SOC). Thus, it is pertinent to identify high-yield OART candidates to best allocate resources. Approach. Reference plans (plans based on simulation anatomy), SOC plans (reference plans recalculated onto daily anatomy), and daily adaptive plans were analyzed for 31 sequential APBI targets, resulting in the analysis of 333 treatment plans. Spearman correlations between 22 reference plan metrics and 10 adaptive benefits, defined as the difference between mean SOC and delivered metrics, were analyzed to select a univariate predictor of OART benefit. A multivariate logistic regression model was then trained to stratify high- and low-benefit candidates. Main results. Adaptively delivered plans showed dosimetric benefit as compared to SOC plans for most plan metrics, although the degree of adaptive benefit varied per patient. The univariate model showed high likelihood for dosimetric adaptive benefit when the reference plan ipsilateral breast V15Gy exceeds 23.5%. Recursive feature elimination identified 5 metrics that predict high-dosimetric-benefit adaptive patients. Using leave-one-out cross validation, the univariate and multivariate models classified targets with 74.2% and 83.9% accuracy, resulting in improvement in per-fraction adaptive benefit between targets identified as high- and low-yield for 7/10 and 8/10 plan metrics, respectively. Significance. This retrospective, exploratory study demonstrated that dosimetric benefit can be predicted using only ipsilateral breast V15Gy on the reference treatment plan, allowing for a simple, interpretable model. Using multivariate logistic regression for adaptive benefit prediction led to increased accuracy at the cost of a more complicated model. This work presents a methodology for clinics wishing to triage OART resource allocation.

Symmetry of constellation diagram-based intelligent SNR estimation for visible light communications.

Visible light communication (VLC) technology with rich spectrum resources is thought of as an essential component in the future ubiquitous communication networks. Accurately monitoring its transmission impairments is important for improving the stability of high-speed communication networks. Existing research on intelligently monitoring the signal-to-noise ratio (SNR) performance of VLC focuses primarily on the application of neural networks but neglects the physical nature of communication systems. In this work, we propose an intelligent SNR estimation scheme for VLC systems, which is based on the symmetry of constellation diagrams with classical deep learning frameworks. In order to increase the accuracy of the SNR estimation scheme, we introduce two data augmentation methods (DA): point normalization and quadrant normalization. The results of extensive simulations demonstrate that the proposed point normalization method is capable of improving accuracy by about 5, 10, 14, and 26%, respectively, for 16-, 64-, 256-, and 1024-quadrature amplitude modulation compared with the same network frameworks without DA. The effect of accuracy improvement can be further superimposed with traditional DA methods. Additionally, the extensive number of constellation points (e.g., 32, 64, 128, 256, 512, 1024, and 2048) on the accuracy of SNR estimation is also investigated.

A recurrent Gaussian quantum network for online processing of quantum time series

Over the last decade, researchers have studied the interplay between quantum computing and classical machine learning algorithms. However, measurements often disturb or destroy quantum states, requiring multiple repetitions of data processing to estimate observable values. In particular, this prevents online (real-time, single-shot) processing of temporal data as measurements are commonly performed during intermediate stages. Recently, it was proposed to sidestep this issue by focusing on tasks with quantum output, eliminating the need for detectors. Inspired by reservoir computers, a model was proposed where only a subset of the internal parameters are trained while keeping the others fixed at random values. Here, we also process quantum time series, but we do so using a Recurrent Gaussian Quantum Network (RGQN) of which all internal interactions can be trained. As expected, this increased flexibility yields higher performance in benchmark tasks. Building on this, we show that the RGQN can tackle two quantum communication tasks, while also removing some hardware restrictions of the currently available methods. First, our approach is more resource efficient to enhance the transmission rate of quantum channels that experience certain memory effects. Second, it can counteract similar memory effects if they are unwanted, a task that could previously only be solved when redundantly encoded input signals could be provided. Finally, we run a small-scale version of the last task on Xanadu’s photonic processor Borealis.

Risk Identification of Mountain Torrent Hazard Using Machine Learning and Bayesian Model Averaging Techniques

Frequent mountain torrent disasters have caused significant losses to human life and wealth security and restricted the economic and social development of mountain areas. Therefore, accurate identification of mountain torrent hazards is crucial for disaster prevention and reduction. In this study, based on historical mountain torrent hazards, a mountain torrent hazard prediction model was established by using Bayesian Model Average (BMA) and three classic machine learning algorithms (gradient-boosted decision tree (GBDT), backpropagation neural network (BP), and random forest (RF)). The mountain torrent hazard condition factors used in modeling were distance to river, elevation, precipitation, slope, gross domestic product (GDP), population, and land use type. Based on the proposed BMA model, flood risk maps were produced using GIS. The results demonstrated that the BMA model significantly improved upon the accuracy and stability of single models in identifying mountain torrent hazards. The F1-values (comprehensively displays the Precision and Recall) of the BMA model under three sets of test samples at different locations were 3.31–24.61% higher than those of single models. The risk assessment results of mountain torrents found that high-risk areas were mainly concentrated in the northern border and southern valleys of Yuanyang County, China. In addition, the feature importance analysis result demonstrated that distance to river and elevation were the most important factors affecting mountain torrent hazards. The construction of projects in mountainous areas should be as far away from rivers and low-lying areas as possible. The results of this study can provide a scientific basis for improving the identification methods of mountain torrent hazards and assisting decision-makers in the implementation of appropriate measures for mountain torrent hazard prevention and reduction.

Diagnosis of Multiple Sclerosis Lesion using Deep Learning Models

Exploring the identification of Multiple Sclerosis (MS) lesions involves leveraging diverse machine learning classifiers. Using Magnetic Resonance Imaging (MRI) scans, the study aims to detect and characterize these lesions, evaluating their attributes, progression stages, and the pivotal role of Artificial Intelligence (AI) in diagnosis. The focus is on analyzing automated detection algorithms, particularly Deep Learning techniques. Through comprehensive assessment, various classifiers including MLP Classifier, Random Forest (RF), Support Vector Machine (SVM), and DL are evaluated using metrics like precision, recall, F1-score, and accuracy. The DL classifier stands out, achieving remarkable precision (99%), recall (99%), and overall accuracy (99.5%). Comparative analysis confirms its superiority, reinforcing its efficacy over alternative methods. The research underscores the DL model's potential in generalizing to new samples due to its robustness and precision. This study significantly advances automated MS lesion detection, highlighting the promise of AI-based methodologies in medical image analysis.

Online System for Monitoring the Degree of Fermentation of Oolong Tea Using Integrated Visible-Near-Infrared Spectroscopy and Image-Processing Technologies.

During the fermentation process of Oolong tea, significant changes occur in both its external characteristics and its internal components. This study aims to determine the fermentation degree of Oolong tea using visible-near-infrared spectroscopy (vis-VIS-NIR) and image processing. The preprocessed vis-VIS-NIR spectral data are fused with image features after sequential projection algorithm (SPA) feature selection. Subsequently, traditional machine learning and deep learning classification models are compared, with the support vector machine (SVM) and convolutional neural network (CNN) models yielding the highest prediction rates among traditional machine learning models and deep learning models with 97.14% and 95.15% in the prediction set, respectively. The results indicate that VIS-NIR combined with image processing possesses the capability for rapid non-destructive online determination of the fermentation degree of Oolong tea. Additionally, the predictive rate of traditional machine learning models exceeds that of deep learning models in this study. This study provides a theoretical basis for the fermentation of Oolong tea.