Supervised Learning Techniques Research Articles

An active learning enhanced data programming (ActDP) framework for ECG time series

Supervised machine learning learns a mapping from input data to output labels, based on the patterns and relationships present in a huge labelled training data.Getting labelled data generally requires a substantial allocation of resources in terms of cost and time. In such scenarios, weak supervised learning techniques like data programming (DP) and active learning (AL) can be advantageous for time-series classification tasks. These paradigms can be used to assign data labels in an automated manner, and time-series classification can subsequently be carried out on the labeled data. This work proposes a novel framework titled AL enhanced data programming (ActDP). It uses a combination of DP and AL for electrocardiogram (ECG) beat classification using single-lead data. ECG beat classification is pivotal in cardiology and healthcare applications for diagnosing a broad spectrum of heart conditions and arrhythmias. To establish the usefulness of this proposed ActDP framework, the experiments have been conducted using the MIT-BIH dataset with 94,224 ECG beats. DP assigns a probabilistic label to each ECG beat using nine novel polar labelling functions and a generative model in this work. Further, AL improves the result of DP by replacing the labels for sampled ECG beats of a generative model with ground truth. Subsequently, a discriminative model is trained on these labels for each iteration. The experimental results show that by incorporating AL into DP in the ActDP framework, the accuracy of ECG classification strictly increases from 85.7% to 97.34% in 58 iterations. Comparatively, the proposed framework (ActDP) has demonstrated a higher classification accuracy of 97.34%. In contrast, DP with data augmentation (DA) achieves an accuracy of 92.2%, while DP without DA results in an accuracy of 85.7%, few-shot learning techniques yield 87.5%–89.2%, and multi-instance learning methods achieve accuracies in the range of 88.9%–94.1%

Real-time AIoT anomaly detection for industrial diesel generator based an efficient deep learning CNN-LSTM in industry 4.0

Anomaly detection for industrial diesel generators, in which unexpected faults could lead to severe consequences, is still challenged due to their complex structure and nonstationary operation. Maintenance engineers who manually audit diesel generators for anomaly detection require significant expertise and knowledge. This study proposes a real-time intelligent AIoT system-based convolution neural network long short-term memory (CNN-LSTM) to enhance efficiency and decrease labor costs of industrial diesel generator maintenance service. The AIoT system could autonomously classify abnormal conditions of industrial diesel generators through supervised learning techniques. Several anomaly failure conditions are identified by maintenance experts and are simulated in the laboratory to collect the working parameters based on developed IoT modules. Pearson product-moment coefficients are computed to effectively evaluate the interdependence between collected variables and the target anomaly types. The proposed CNN-LSTM structure is hyperparameter fine-tuning for identifying the most critical configurations in failure-diagnosing applications. The developed approach is comprehensively analyzed and evaluated with other state-of-the-art individual deep learning algorithms, including recurrent neural network (RNN), LSTM, gate-recurrent unit (GRU), and CNN. The experiment results indicate that the proposed hybrid CNN-LSTM could achieve distinguished diagnosis precision of anomaly conditions of industrial diesel generators and significantly improve the classified performance in Industry 4.0.

The effect of different types of comparative reviews on product sales

Comparative online reviews have evolved into a vital instrument for consumers in decision-making, offering valuable comparisons and available options. Drawing on the insights from the linguistic category model (LCM) and elaboration likelihood model (ELM), we propose that different types (attribute-based and experience-based) of comparative reviews can affect consumers' perceived credibility of online reviews, thus impacting product sales. We analyzed 136,260 reviews on e-commerce platforms to assess these effects and introduced review valence as a boundary condition. Utilizing a combination of pattern discovery, supervised learning techniques, and manual coding, we identified attribute-based and experience-based comparative reviews and subsequently classified them based on positive, neutral, and negative valence. Subsequently, we took the product sales as the dependent variable and applied a two-way fixed effects model. The results indicate that attribute-based comparative reviews exert a more favorable influence on product sales compared to experience-based ones. Additionally, positive comparative reviews, irrespective of their attribute-based or experience-based nature, demonstrate a greater impact than regular positive reviews. However, negative and neutral comparative reviews, only when associated with attribute-based information, exhibit a significant effect. The results highlight the value of different types of comparative reviews and illuminate the moderating role of review valence. Our findings offer new insights and practical guidance for marketers and e-commerce platforms in capitalizing on the important influence of comparative reviews and enhancing the presentation of online reviews.

Attention decoupled contrastive learning for semi-supervised segmentation method based on data augmentation

Objective. Deep learning algorithms have demonstrated impressive performance by leveraging large labeled data. However, acquiring pixel-level annotations for medical image analysis, especially in segmentation tasks, is both costly and time-consuming, posing challenges for supervised learning techniques. Existing semi-supervised methods tend to underutilize representations of unlabeled data and handle labeled and unlabeled data separately, neglecting their interdependencies. Approach. To address this issue, we introduce the Data-Augmented Attention-Decoupled Contrastive model (DADC). This model incorporates an attention decoupling module and utilizes contrastive learning to effectively distinguish foreground and background, significantly improving segmentation accuracy. Our approach integrates an augmentation technique that merges information from both labeled and unlabeled data, notably boosting network performance, especially in scenarios with limited labeled data. Main results. We conducted comprehensive experiments on the automated breast ultrasound (ABUS) dataset and the results demonstrate that DADC outperforms existing segmentation methods in terms of segmentation performance.

Cognitive driven gait freezing phase detection and classification for neuro-rehabilitated patients using machine learning algorithms

BackgroundThe significance of diagnosing illnesses associated with brain cognitive and gait freezing phase patterns has led to a recent surge in interest in the study of gait for mental disorders. A more precise and effective way to characterize and classify many common gait problems, such as foot and brain pulse disorders, can improve prognosis evaluation and treatment options for Parkinson patients. Nonetheless, the primary clinical technique for assessing gait abnormalities at the moment is visual inspection, which depends on the subjectivity of the observer and can be inaccurate. Research questionThis study investigates whether it is possible to differentiate between gait brain disorder and the typical walking pattern using machine learning driven supervised learning techniques and data obtained from inertial measurement unit sensors for brain, hip and leg rehabilitation. MethodThe proposed method makes use of the Daphnet freezing of Gait Data Set, consisted of 237 instances with 9 attributes. The method utilizes machine learning and feature reduction approaches in leg and hip gait recognition. ResultsFrom the obtained results, it is concluded that among all classifiers RF achieved highest accuracy as 98.9 % and Perceptron achieved lowest i.e. 70.4 % accuracy. While utilizing LDA as feature reduction approach, KNN, RF and NB also achieved promising accuracy and F1-score in comparison with SVM and LR classifiers. SignificanceIn order to distinguish between the different gait disorders associated with brain tissues freezing/non-freezing and normal walking gait patterns, it is shown that the integration of different machine learning algorithms offers a viable and prospective solution. This research implies the need for an impartial approach to support clinical judgment.

Performance evaluation of BLDC motor drive mounted in aerial vehicle (drone) using adaptive neuro-fuzzy

The development of autonomous drones equipped with cameras and various sensors has paved the way for their application in agriculture and perimeter security. These aerial drones require specific power, acceleration, high torque, and efficiency to meet the demands of agricultural tasks, utilizing built-in brushless DC (BLDC) motors. However, a common challenge drone’s face is maintaining the desired speed for extended periods. Enhancing the performance of BLDC motors through advanced controllers is crucial to address this issue. This research paper proposes optimizing the size and speed of brushless DC motors for aerial vehicles using an adaptive fuzzy inference system and supervised learning techniques. When these drones carry loads, the BLDC motors must dynamically adjust the drone's speed. During this phase, the motors must control their speed and torque using artificial intelligence controllers like adaptive neuro-fuzzy inference systems (ANFIS) to enhance the drone's functionality, resilience, and safety. This research has conducted analyses focused on improving the performance of BLDC motors explicitly personalized for unmanned aerial vehicle (UAVs). The proposed method will be implemented using MATLAB/Simulink, expecting to significantly enhance the BLDC motor's performance compared to conventional controllers. Comparative analyses will be conducted between traditional and ANFIS controllers to validate the effectiveness of the proposed approach.

Supervised learning techniques for blood product prediction in patients with hematologic diseases: a multi-centre study in Western Algeria

Supervised learning techniques for blood product prediction in patients with hematologic diseases: a multi-centre study in Western Algeria

Fake News Detection

The spread of false information on the internet has become a major social issue, casting doubt on the veracity of information shared on these platforms. This study uses cutting-edge methods from machine learning (ML) and natural language processing (NLP) to present a complete framework for the detection of fake news. The purpose of this paper is to develop a model for detecting bogus news. A model is selected by using supervised learning techniques. In addition, we categorize news stories as real or fraudulent using the Naïve Bayes, Logistic Regression, and Random Forest algorithms. Our methodology offers an approach to false news identification that is more robust by taking into account the credibility of the news sources in addition to the content of the news. Using labeled datasets of fictitious and authentic news stories, we train our algorithms. A few methodologies were compared to achieve varying degrees of accuracy. When compared to the other two models, Random Forest is thought to have produced the best results in terms of accuracy. We assess our framework's effectiveness using real-world news articles and benchmark datasets, showcasing its versatility in correctly recognizing false information in a variety of settings and domains. We demonstrate the advantages of our method in terms of detection accuracy, scalability, and computational efficiency by comprehensive experimentation and comparative analysis. All things considered, our suggested framework is a major step forward in the fight against the dissemination of false information on the internet and provides a workable way to lessen the negative effects of fake news on people, communities, and society at large.

Exploring Machine Learning Algorithms for Enhanced Diagnosis of Breast Cancer : A Comparative Study

Breast cancer remains a significant global health concern, necessitating accurate and efficient diagnostic methods for timely intervention and treatment. This study investigates the efficacy of various machine learning algorithms in diagnosing breast cancer based on clinical data. Leveraging a comprehensive dataset comprising demographic information, medical history, and diagnostic features, we employ supervised learning techniques to train and evaluate multiple classifiers. Through a comparative analysis, we assess the performance of popular machine learning algorithms including logistic regression, support vector machines, decision trees, random forests, and neural networks. Evaluation metrics such as accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC-ROC) are utilized to quantify the diagnostic capabilities of each model. Our results demonstrate promising performance across the evaluated algorithms, with some exhibiting superior accuracy and predictive power compared to others. Furthermore, we explore feature importance to gain insights into the characteristics influencing the classification process. This research contributes to the growing body of literature on utilizing machine learning for medical diagnosis, offering valuable insights for developing robust and accurate tools for detecting breast cancer.

Structural Optimization of Trusses in Building Information Modeling (BIM) Projects Using Visual Programming, Evolutionary Algorithms, and Life Cycle Assessment (LCA) Tools

The optimal structural design is imperative in order to minimize material consumption and reduce the environmental impacts of construction. Given the complexity in the formulation of structural design problems, the process of optimization is commonly performed using artificial intelligence (AI) global optimization, such as the genetic algorithm (GA). However, the integration of AI-based optimization, together with visual programming (VP), in building information modeling (BIM) projects warrants further investigation. This study proposes a workflow by combining structure analysis, VP, BIM, and GA to optimize trusses. The methodology encompasses several steps, including the following: (i) generation of parametric trusses in Dynamo VP; (ii) performing finite element modeling (FEM) using Robot Structural Analysis (RSA); (iii) retrieving and evaluating the FEM results interchangeably between Dynamo and RSA; (iv) finding the best solution using GA; and (v) importing the optimized model into Revit, enabling the user to perform simulations and engineering analysis, such as life cycle assessment (LCA) and quantity surveying. This methodology provides a new interoperable framework with minimal interference with existing supply-chain processes, and it will be flexible to technology literacy and allow architectural, engineering and construction (AEC) professionals to employ VP, global optimization, and FEM in BIM-based projects by leveraging open-sourced software and tools, together with commonly used design software. The feasibility of the proposed workflow was tested on benchmark problems and compared with the open literature. The outcomes of this study offer insight into the opportunities and limitations of combining VP, GA, FEA, and BIM for structural optimization applications, particularly to enhance structural efficiency and sustainability in construction. Despite the success of this study in developing a workable, user-friendly, and interoperable framework for the utilization of VP, GA, FEM, and BIM for structural optimization, the results obtained could be improved by (i) increasing the callback function speed between Dynamo and RSA through specialized application programming interface (API); and (ii) fine-tuning the GA parameters or utilizing other advanced global optimization and supervised learning techniques for the optimization.

Standby Power Reduction of Home Appliance by the i-HEMS System Using Supervised Learning Techniques

Electricity consumption in homes is on the rise due to the increasing prevalence of home appliances and longer hours spent indoors. Home energy management systems (HEMSs) are emerging as a solution to reduce electricity consumption and efficiently manage power usage at home. In the past, numerous studies have been conducted on the management of electricity production and consumption through solar power. However, there are limited human-centered studies focusing on the user’s lifestyle. In this study, we propose an Intelligent Home Energy Management System (i-HEMS) and evaluate its energy-saving effectiveness through a demonstration in a standard house in Korea. The system utilizes an IoT environment, PID sensing, and behavioral pattern algorithms. We developed algorithms based on power usage monitoring data of home appliances and human body detection. These algorithms are used as the primary scheduling algorithm and a secondary algorithm for backup purposes. We explored the deep connection between power usage, environmental sensor data, and input schedule data based on Long Short-Term Memory network (LSTM) and developed an occupancy prediction algorithm. We analyzed the use of common home appliances (TV, computer, water purifier, microwave, washing machine, etc.) in a standard house and the power consumption reduction by the i-HEMS system. Through a total of six days of empirical experiments, before implementing i-HEMS, home appliances consumed 13,062 Wh. With i-HEMS, the total consumption was reduced to 10,434 Wh (a 20% reduction), with 9060 Wh attributed to home appliances and 1374 Wh to i-HEMS operation.

Contrastive learning based open-set recognition with unknown score

Open-set recognition (OSR) is designed to classify the seen classes and identify the unseen classes as unknown. However, existing open-set classifiers rely on deep networks trained using supervised learning techniques for known classes present in the training samples. This factor tends to specialize the learned features of the model toward known classes, making it challenging to differentiate between unknown classes. In this study, we propose a novel open-set recognition framework based on contrastive learning with an unknown score. First, we propose a novel training framework based on contrastive learning to learn more informative features and preserve beneficial information to separate unknown from known. Second, we propose an unknown score function based on multi-layer features to detect unknown samples by considering the difference between the known and unknown classes of intermediate-layer features. Based on extensive experiments conducted on multiple benchmark datasets, the proposed method has been demonstrated to outperform existing methods and achieve state-of-the-art results.

Shifting to machine supervision: annotation-efficient semi and self-supervised learning for automatic medical image segmentation and classification

Advancements in clinical treatment are increasingly constrained by the limitations of supervised learning techniques, which depend heavily on large volumes of annotated data. The annotation process is not only costly but also demands substantial time from clinical specialists. Addressing this issue, we introduce the S4MI (Self-Supervision and Semi-Supervision for Medical Imaging) pipeline, a novel approach that leverages advancements in self-supervised and semi-supervised learning. These techniques engage in auxiliary tasks that do not require labeling, thus simplifying the scaling of machine supervision compared to fully-supervised methods. Our study benchmarks these techniques on three distinct medical imaging datasets to evaluate their effectiveness in classification and segmentation tasks. Notably, we observed that self-supervised learning significantly surpassed the performance of supervised methods in the classification of all evaluated datasets. Remarkably, the semi-supervised approach demonstrated superior outcomes in segmentation, outperforming fully-supervised methods while using 50% fewer labels across all datasets. In line with our commitment to contributing to the scientific community, we have made the S4MI code openly accessible, allowing for broader application and further development of these methods. The code can be accessed at https://github.com/pranavsinghps1/S4MI.

Agent Based Intelligent System for Enhanced Teamwork Performance

It is impossible to overstate the necessity of a strategic and practical approach in the workplace in order to maximize productivity these days. Teamwork is one of the best ways to adapt to the changes that have occurred in today's environment throughout time. In every industry, the optimum performance arrangement for realizing visions, carrying out plans, and accomplishing objectives is teamwork. It is also one of the most crucial components of systems for continuous improvement since it makes information exchange, issue resolution, and the growth of employee accountability easier. Teams function as a grouping of people with complementary talents who work together rather than against one another. They are held accountable for their strategic methods and use them to achieve a shared objective. The Supervised Learning technique was used in this work to simulate team performance utilizing an intelligent coaching agent. Through the use of an automated performance assessment and weighted scores for each task, this study was able to create a system that will remove biases from performance evaluation. As soon as a worker does the task, they will obtain a score. The purpose of this study was to demonstrate an event-based performance approach by developing and utilizing an intelligent coaching agent in a supervised learning team training framework. The goal was successfully met, and the result shows positive impacts on the team's performance.

Deployment of a Phenotypic Characterization System for Effective Identification of the Onset of Asthma Disease

Background Essentially, machine learning techniques help with clinical decision-making by forecasting prediction results based on recent and historical data, which are frequently found in carefully chosen clinical data repositories. In order to uncover hidden patterns in the data, machine learning applies sophisticated analytical techniques that conduct an exploratory analysis while constructing prediction models to support clinical judgment. Objective To effectively identify asthmatics in two distinct cohorts representing India's rural and urban populations by adopting a phenotypic characterization approach. Methods Cross-sectional and categorical in design, the data represent the two populations, with clinical history information emphasizing clinical symptoms and patterns defining the condition. The method adopts a hybrid approach since it uniquely blends the unsupervised and supervised learning techniques to explore the advantages of both. The clustering data emphasizing the phenotypic characteristics of asthma is input to the classifier, and the performance of the classifier was continuously monitored for significant improvement in the results. Results Asthma disease outcome predictions made by the hybrid decision support system were quite accurate, with classification accuracy reaching up to 85.1% and 95.3% for the two datasets, respectively. Conclusion Since asthma is a heterogeneous disease with multiple subtypes, employing clustering information in the form of cluster evaluation scores as an input parameter to the classifiers can effectively predict disease outcomes.

Learned load balancing

Effectively balancing traffic in datacenter networks is a crucial operational goal. Most existing load balancing approaches are handcrafted to the structure of the network and/or network workloads. Thus, new load balancing strategies are required if the underlying network conditions change, e.g., due to hard or grey failures, network topology evolution, or workload shifts. While we can theoretically derive the optimal load balancing strategy by solving an optimization problem given certain traffic and topology conditions, these problems take too much time to solve and makes the derived solution stale to deploy. In this paper, we describe a load balancing scheme Learned Load Balancing (LLB), which is a general approach to finding an optimal load balancing strategy for a given network topology and workload, and is fast enough in practice to deploy the inferred strategies. LLB uses deep supervised learning techniques to learn how to handle different traffic patterns and topology changes, and adapts to any failures in the underlying network. LLB leverages emerging trends in network telemetry, programmable switching, and “smart” NICs. Our experiments show that LLB performs well under failures and can be expanded to more complex, multi-layered network topologies. We also prototype neural network inference on smartNICs to demonstrate the workability of LLB.

Supervised and unsupervised techniques to analyze risk factors associated with motorcycle crash.

Motorcycles are one of the highly used modes of transport in Barcelona, Spain, in particular, and in many different regions, in general. This situation is compromising safety on the road and may be attributed to a potential increase in traffic crashes. Therefore, this study examines several risk factors and their consequential impacts on the level of injury that is resulted in case of a traffic crash. Two approaches are employed to analyze the risk factors, including a supervised learning technique which is a binary probit model, and an unsupervised technique which is the Kohonen clustering. The results for both models show that alcoholism and road in poor condition can indeed increase the probability of having different levels of injuries as reasons for the crash. Elderly users are less likely to be involved in motorcycle crash injuries compared to other age categories, especially the age group that ranges from 25 to 40 years old which has the highest odds. For both techniques, the performance in analyzing the utilized data shows that both approaches can be successfully utilized for this type of dataset. This study highlights the considerable danger faced by motorcyclists due to various risk factors. It stresses the critical importance of maintaining roads in optimal condition not just for efficient travel but also to enhance motorcyclists' safety. Additionally, the research underscores the significant threat posed by speeding, particularly exceeding speed limits, to motorcyclists' safety, emphasizing the urgent need for more 30 km/h speed limit zones and stricter enforcement of speed regulations. As a result, the research has identified several risk factors that increase the likelihood of severe or fatal injuries among motorcyclists in Barcelona and has suggested certain recommendations to mitigate their impact.

Phishing Email Detection using Machine Learning

The increasing sophistication of phishing attacks poses a significant threat to cybersecurity, demanding advanced and adaptive detection mechanisms. This project focuses on the development and implementation of a phishing detection system leveraging machine learning algorithms. The objective is to enhance the security measures for users by employing a robust and intelligent system capable of identifying and thwarting phishing attempts. The project involves the collection and curation of diverse datasets containing examples of phishing and legitimate communication. Supervised learning techniques will be employed to train the machine learning model, utilizing features such as email content, sender information, and website attributes. The model's training will be continuously refined to adapt to emerging phishing tactics and enhance its accuracy. Additionally, unsupervised learning methods will be explored to detect novel phishing patterns and anomalies that may not be present in the training dataset. Clustering algorithms will assist in grouping similar instances, aiding the identification of previously unseen phishing techniques. The implementation will be evaluated through extensive testing using real-world datasets, measuring the system's effectiveness in detecting phishing attempts with a low false-positive rate. The project aims to contribute to the field of cybersecurity by providing a proactive and adaptive solution to counter the evolving nature of phishing threats. The successful implementation of this phishing detection system will enhance user confidence in online interactions and contribute to the overall resilience of digital ecosystems. Key Words: Phishing attacks, Cybersecurity, Detection mechanisms, Machine learning algorithms, Supervised learning, Unsupervised learning, Email content analysis, Sender information, Website attributes, Training dataset, Real-world datasets, False-positive rate, Proactive cybersecurity, Adaptive solutions, Digital ecosystems.

ACCIDENT DETECTION USING TENSOR FLOW AND CNN

Accident detection is a critical aspect of road safety management, with timely identification and response being paramount to minimizing casualties and damage. In this project, a comprehensive approach to accident detection utilizing Deep learning techniques is proposed. The system aims to automatically detect accidents in real time using information obtained from diverse sources like video feeds, sensor inputs, and audio signals. The project involves the development and implementation of Deep learning models trained on labeled datasets containing examples of accident and non-accident scenarios. Supervised learning techniques like classification, regression and advanced neural network architecture like convolution, recurrent neural network are considered to effectively process and analyze the data. The deployed system continuously monitors the environment, processing incoming data streams to identify patterns indicative of accidents. Upon detection, the system triggers appropriate response mechanisms, such as alerting emergency services or nearby vehicles, facilitating swift intervention. The project aims to contribute to road safety initiatives by providing a scalable and efficient solution for accident detection. Additionally, the insights gained from the data collected by the system can inform policymakers and urban planners in implementing measures to prevent accidents and improve overall road safety. Through this project, the students gain valuable experience in applying Deep learning techniques to address real-world challenges and contribute to societal welfare.

Applying machine learning to assess emotional reactions to video game content streamed on Spanish Twitch channels

This research explores for the first time the application of machine learning to detect emotional responses in video game streaming channels, specifically on Twitch, the most widely used platform for broadcasting content. Analyzing sentiment in gaming contexts is difficult due to the brevity of messages, the lack of context, and the use of informal language, which is exacerbated in the gaming environment by slang, abbreviations, memes, and jargon. First, a novel Spanish corpus was created from chat messages on Spanish video game Twitch channels, manually labeled for polarity and emotions. It is noteworthy as the first Spanish corpus for analyzing social responses on Twitch. Secondly, machine learning algorithms were used to classify polarity and emotions offering promising evaluations. The methodology followed in this work consists of three main steps: (1) Extracting Twitch chat messages from Spanish streamers’ channels related to gaming events and gameplays; (2) Processing and selecting the messages to form the corpus and manually annotating polarity and emotions; and (3) Applying machine learning models to detect polarity and emotions in the created corpus. The results have shown that a Bidirectional Encoder Representation from Transformers (BERT) based model excels with 78% accuracy in polarity detection, while deep learning and Random Forest models reach around 70%. For emotion detection, the BERT model performs best with 68%, followed by deep learning with 55%. It is worth noting that emotion detection is more challenging due to the subjective interpretation of emotions in the complex communicative context of video gaming on platforms such as Twitch. The use of supervised learning techniques, together with the rigorous corpus labeling process and the subsequent corpus pre-processing methodology, has helped to mitigate these challenges, and the algorithms have performed well. The main limitations of the research involve category and video game representation balance. Finally, it is important to stress that the integration of machine learning in video games and on Twitch is innovative, by allowing the identification of viewers’ emotions on streamers’ channels. This innovation could bring benefits such as a better understanding of audience sentiment, improving content and audience retention, providing personalized recommendations and detecting toxic behavior in chats.