High Accuracy Rate Research Articles

Zero-shot Learning for Subdiscrimination in Pre-trained Models

In deep metric learning (DML) high-level input data are represented in a lower-level representation (embedding) space, such that samples from the same class are mapped close together, while samples from disparate classes are mapped further apart. In this lower-level representation, only a single inference sample from each known class is required to accurately discriminate between classes. To this end, embeddings trained for a specific task may contain additional feature information which can be used to go a level deeper into the discrimination task, i.e. allowing for feature sub-discrimination. This study takes an embedding trained to discriminate faces (identities) and uses the inherent feature information within the embedding to differentiate several attributes such as gender, age, and skin tone, without any additional training. This study is split into two cases; intra class discrimination where all the embeddings considered are from the same identity/in-dividual but with minor attributes such as beard/beardless, glasses/without glasses and emotions; and extra class discrimination where the embeddings represent different identities/people with more prominent attributes such as male/female, pale/dark tone, young/older. In the intra class sub-discriminant scenario, the inference process distinguishes common attributes and several artefacts of different identities, achieving 90.0% and 76.0% accuracy for beards and glasses, respectively. The system can also perform extra class sub-discrimination with a high accuracy rate, notably 99.3%, 99.3% and 94.1% for gender, skin tone, and age, respectively. To sum up, this work investigates the sub-discriminative capabilities of DML models by clustering discriminative features evident within the structure of DML embeddings.

Classification of CT scan and X-ray dataset based on deep learning and particle swarm optimization.

In 2019, the novel coronavirus swept the world, exposing the monitoring and early warning problems of the medical system. Computer-aided diagnosis models based on deep learning have good universality and can well alleviate these problems. However, traditional image processing methods may lead to high false positive rates, which is unacceptable in disease monitoring and early warning. This paper proposes a low false positive rate disease detection method based on COVID-19 lung images and establishes a two-stage optimization model. In the first stage, the model is trained using classical gradient descent, and relevant features are extracted; in the second stage, an objective function that minimizes the false positive rate is constructed to obtain a network model with high accuracy and low false positive rate. Therefore, the proposed method has the potential to effectively classify medical images. The proposed model was verified using a public COVID-19 radiology dataset and a public COVID-19 lung CT scan dataset. The results show that the model has made significant progress, with the false positive rate reduced to 11.3% and 7.5%, and the area under the ROC curve increased to 92.8% and 97.01%.

Prediction of surgical necessity in children with ureteropelvic junction obstruction using machine learning.

Hydronephrosis developing at the ureteropelvic junction due to obstruction poses clinical challenges as it has the potential to cause renal damage. This study aims to evaluate how well machine learning models such, as XGBClassifier and Logistic Regression can be used to predict the need for treatment in patients, with hydronephrosis resulting from ureteropelvic junction obstruction. Hydronephrosis was diagnosed in the medical records of patients from January 2015 to December 2020. These patients were classified into two groups: those who were not operated upon (n = 194) and those who had surgical procedures (n = 129). Details such as demographics, clinical presentations, and imaging findings were captured. XGBClassifier and Logistic Regression methods were employed to predict the requirement for an operation. The performance of the models was assessed based on ROC-AUC values, sensitivity, and specificity. The XGBClassifier algorithm gave the best prediction results with a ROC-AUC value of 0.977 and an accuracy rate of 95.4%. The Logistic Regression algorithm, on the other hand, offered the highest prediction during cross-validation. The presence of obstruction on scintigraphy, kidney size, anteroposterior diameter of the renal pelvic and parenchymal thickness observed in hydronephrotic kidney on USG have been identified as important predictive factors. In predicting the requirement for surgery in cases of hydronephrosis due to obstruction, machine learning algorithms have shown high accuracy and sensitivity rates. Consequently, clinical decision support systems based on these algorithms may lead to better care management of patients and more accurate projections concerning the need for surgical intervention. ESH/GOEK 2024/88-23/01/2024.

Automated Classification of Biscuit Quality Using YOLOv8 Models in Food Industry

Abstract It is of great importance for food safety and consumer satisfaction that industrial food products are durable, hygienic, and flawless. Robust products protect the physical integrity of the product by preventing damage that may occur during the production and transportation processes, which meets the expectations of the consumer. Hygienic production conditions prevent foodborne diseases by minimizing the risk of microbial contamination and protect consumer health. Perfect products strengthen the brand image with their aesthetic and satisfactory features and increase consumer loyalty. In the study conducted in this context, the classification of defect and no defect conditions of biscuits in the food industry was examined using YOLOv8 models. A summary dataset consisting of 4990 biscuit images was created and the biscuits were initially divided into two categories: defect and no defect. Later, defect biscuits were classified into three subcategories: not complete, overcooked, and texture defect. As a result of experiments with YOLOv8 models, binary classification (defect, no defect), the highest accuracy rate was achieved in the YOLOv8-m, YOLOv8-l, and YOLOv8-x models with 96.78%, while the highest accuracy rate in the triple classification (not complete, overcooked, and texture defect) performance was achieved in the YOLOv8-m model with 96.99%.

The joint recognition of multi-MHD instabilities on HL-2A

Abstract In tokamak plasmas, various Magnetohydrodynamic (MHD) instabilities could be driven by free energy, which enhance the plasma's transportation, leading to a reduction in critical fusion parameters such as temperature and density, and in severe cases, can even cause major plasma disruptions.This study employs deep learning techniques to learn from 1000 shots manually labelled as three types of MHD instabilities: fishbone mode, long-lived mode, and tearing mode, enabling real-time automated recognition of the instabilities. High accuracies of 97.83%, 95.32%, 94.84% are obtained on 200 testing shots, which are measured by area under the receiver-operator characteristic curve (AUC). Data processing methods that conform to the intuition of physics experts, such as Short Time Fourier Transform (STFT), have been retained, and advanced Artificial Intelligence (AI) algorithms such as Resnet have been combined to achieve a high accuracy rate. It also demonstrated the robustness in fully automatic detections over thousands of discharges. Furthermore, this study explores multitask learning techniques. Instead of using three individual neural network to recognize the different instabilities, a joint recognition algorithm is proposed. The joint algorithm shares the encoder of the three networks and use separate decoder branches to output the result of different instabilities. An inspiring outcome is found that the joint algorithm outperforms the individual ones on all of the instability recognition tasks. Implementing multiple MHD recognition tasks jointly can comprehensively improve the model's performance on each task by sharing related information between intrinsically related tasks. This means that in the future, the model can further develop as more tasks are added, revealing a possible technique routine to build an accurate and comprehensive large-scale model for fusion applications.

Wood-waste classification using interpretable machine learning

ABSTRACT Determining and classifying the contents of wood wastes, which possess an extremely heterogeneous structure, is crucial for their optimal utilization. The heterogeneous nature of waste wood poses significant challenges for its separation and classification. Also, the classical chemometric methods are not ergonomic in terms of time and cost. For this reason, this study aimed to develop an efficient ML-based decision support system (DSS) for accurate classification of waste wood in addition to being an ergonomic approach versus chemometric methods. To develop ML-based DSS, firstly absorbance values at 650–4000 cm−1 wavelength were measured using FTIR-ATR device for 200 wood-waste samples. Absorbance values at 52 key wavelengths, identified as the most effective for characterizing the samples, were used as input features to estimate ML classification models using artificial neural networks (ANN), random forest (RF), ensemble learning, multiple logistic regression (MLR), discriminant analysis, Naive Bayes, K-nearest neighbour (KNN), and support vector machines (SVM). The analysis results indicated that the developed DSS achieved quite a high classification accuracy rates, approaching 100%. As a result, this kind of ML-based DSS can be used as a practical and efficient tool to determine and classify the contents of wood wastes.

Diagnosis of Reverse-Connection Defects in High-Voltage Cable Cross-Bonded Grounding System Based on ARO-SVM.

High-voltage (HV) cables are increasingly used in urban power grids, and their safe operation is critical to grid stability. Previous studies have analyzed various defects, including the open circuit in the sheath loop, the flooding in the cross-bonded link box, and the sheath grounding fault. However, there is a paucity of research on the defect of the reverse direction between the inner core and the outer shield of the coaxial cable. Firstly, this paper performed a theoretical analysis of the sheath current in the reversed-connection state and established a simulation model for verification. The outcomes of the simulation demonstrate that there are significant variations in the amplitudes of the sheath current under different reversed-connection conditions. Consequently, a feature vector was devised based on the amplitude of the sheath current. The support vector machine (SVM) was then applied to diagnose the reversed-connection defects in the HV cable cross-bonded grounding system. The artificial rabbits optimization (ARO) algorithm was adopted to optimize the SVM model, attaining an impressively high diagnostic accuracy rate of 99.35%. The effectiveness and feasibility of the proposed algorithm are confirmed through the analysis and validation of the practical example.

Fraud Detection in Banking using Key Agreement and Face Authentication

Fraud detection in banking is a critical domain for financial security, driven by the rising frequency and sophistication of cyber-attacks. This paper introduces a dual-layered fraud detection framework combining key agreement protocols and face authentication. The system ensures secure communication through cryptographic techniques, such as Diffie-Hellman and Elliptic Curve Cryptography, while leveraging real-time facial recognition powered by TensorFlow and OpenCV for biometric authentication. These technologies collectively address identity theft, unauthorized access, and data breaches in banking operations. Implementation results demonstrate a high accuracy rate in user verification, efficient transaction processing, and robust protection against cyber threats. This paper outlines the system's architecture, implementation, and performance while addressing associated challenges such as privacy concerns, algorithmic biases, and scalability. The findings highlight the transformative potential of integrating cryptography and biometric authentication for modern banking security.

Evaluation of different artificial intelligence applications in responding to regenerative endodontic procedures

IntroductionThe integration of artificial intelligence (AI) technologies in healthcare is revolutionizing the workflows of healthcare professionals, enabling faster and more accurate patient treatment. This study aims to evaluate the accuracy of responses provided by different AI chatbots to questions that dentists might ask regarding regenerative endodontic treatment (RET), a procedure that shows promising biological healing potential.MethodsA total of 23 questions related to RET procedures were developed based on the American Association of Endodontists (AAE) 2022 guidelines. These questions were presented to three platforms: Google Bard (Gemini), ChatGPT-4o, and ChatGPT-4 with a PDF plugin. The questions were asked three times a day over the course of 10 days by two different researchers. The responses obtained were categorized as correct, incorrect, or insufficient according to the guidelines, and the results were statistically analyzed.ResultsA total of 1,380 responses were collected from the three platforms. Statistical analysis revealed significant relationships between the responses given by the platforms during morning, noon, and evening sessions (p < 0.05). ChatGPT-4 with the PDF plugin had the highest correct response rate (98.1%), while Gemini depicted the lowest (48%).ConclusionsWhile ChatGPT-4o and Gemini offer promising results for regenerative endodontic clinical applications, they do not provide sufficient guidance. However, ChatGPT-4 with the PDF plugin could be a valuable tool for clinicians due to its high accuracy rate. Further research is needed to develop AI programs specifically designed for clinicians in the field of endodontics.

Extracting Fruit Disease Knowledge from Research Papers Based on Large Language Models and Prompt Engineering

In China, fruit tree diseases are a significant threat to the development of the fruit tree industry, and knowledge about fruit tree diseases is the most needed professional knowledge for fruit farmers and other practitioners in the fruit tree industry. Research papers are the primary sources of professional knowledge that represent the cutting-edge progress in fruit disease research. Traditional knowledge engineering methods for knowledge acquisition require extensive and cumbersome preparatory work, and they demand a high level of professional background and information technology skills from the handlers. This paper, from the perspective of fruit tree industry knowledge dissemination, aims at users such as fruit farmers, fruit tree experts, fruit tree knowledge communicators, and information gatherers. It proposes a fast, cost-effective, and low-technical-barrier method for extracting fruit tree disease knowledge from research paper abstracts—K-Extract, based on large language models (LLMs) and prompt engineering. Under zero-shot conditions, K-Extract utilizes conversational LLMs to automate the extraction of fruit tree disease knowledge. The K-Extract method has constructed a comprehensive classification system for fruit tree diseases and, through a series of optimized prompt questions, effectively overcomes the deficiencies of LLM models in providing factual accuracy. This paper tests multiple LLM models available in the Chinese market, and the results show that K-Extract can seamlessly integrate with any conversational LLM model, with the DeepSeek model and the Kimi model performing particularly well. The experimental results indicate that LLM models have a high accuracy rate in handling judgment tasks and simple knowledge Q&amp;A tasks. The K-Extract method is simple, efficient, and accurate, and can serve as a convenient tool for knowledge extraction in the agricultural field.

Parentage Verification and Segregation Distortion Patterns of Microsatellite Markers in Olive Flounder (Paralichthys olivaceus) Full-Sib Families.

Microsatellite markers are widely used in aquaculture for genetic analysis and breeding programs, but challenges such as segregation distortion and allelic instability can impact their effectiveness in parentage verification and inheritance studies. This study evaluated 15 microsatellite loci in seven experimental olive flounder (Paralichthys olivaceus) families bred through 1:1 full-sibling crosses, assessing their utility for accurate parentage and inheritance stability. Parentage assignments were conducted within an expanded pool of 647 candidate parents (including the actual 14 parents), encompassing both closely related and moderately distant individuals. Despite increased genetic diversity, assignments maintained a high accuracy rate (99.6%), demonstrating marker robustness. Differences in delta values highlighted the influence of genetic backgrounds on assignment clarity, with some groups showing reduced distinctiveness in the expanded pool. Segregation distortion was observed at certain loci, deviating from Mendelian inheritance, likely due to meiotic drive and post-zygotic selection. These findings underscore the need for empirical validation of microsatellite loci for stable inheritance and reliable parentage in diverse breeding programs, especially with genetically similar spawners.

Leveraging Deep Learning for Real-Time Coffee Leaf Disease Identification

Agriculture is vital for providing food and economic benefits, but crop diseases pose significant challenges, including coffee cultivation. Traditional methods for disease identification are labor-intensive and lack real-time capabilities. This study aims to address existing methods’ limitations and provide a more efficient, reliable, and cost-effective solution for coffee leaf disease identification. It presents a novel approach to the real-time identification of coffee leaf diseases using deep learning. We implemented several transfer learning (TL) models, including ResNet101, Xception, CoffNet, and VGG16, to evaluate the feasibility and reliability of our solution. The experiment results show that the proposed models achieved high accuracy rates of 97.30%, 97.60%, 97.88%, and 99.89%, respectively. CoffNet, our proposed model, showed a notable processing speed of 125.93 frames per second (fps), making it suitable for real-time applications. Using a diverse dataset of mixed images from multiple devices, our approach reduces the workload of farmers and simplifies the disease detection process. The findings lay the groundwork for the development of practical and efficient systems that can assist coffee growers in disease management, promoting sustainable farming practices, and food security.

A Diagnostic and Performance System for Soccer: Technical Design and Development.

This study presents a novel system for diagnosing and evaluating soccer performance using wearable inertial sensors integrated into players' insoles. Designed to meet the needs of professional podiatrists and sports practitioners, the system focuses on three key soccer-related movements: passing, shooting, and changes of direction (CoDs). The system leverages low-power IMU sensors, Bluetooth Low Energy (BLE) communication, and a cloud-based architecture to enable real-time data analysis and performance feedback. Data were collected from nine professional players from the SD Huesca women's team during controlled tests, and bespoke algorithms were developed to process kinematic data for precise event detection. Results indicate high accuracy rates for detecting ball-striking events and CoDs, with improvements in algorithm performance achieved through adaptive thresholds and ensemble neural network models. Compared to existing systems, this approach significantly reduces costs and enhances practicality by minimizing the number of sensors required while ensuring real-time evaluation capabilities. However, the study is limited by a small sample size, which restricts generalizability. Future research will aim to expand the dataset, include diverse sports, and integrate additional sensors for broader applications. This system offers a valuable tool for injury prevention, player rehabilitation, and performance optimization in professional soccer, bridging technical advancements with practical applications in sports science.

Classification of Cardiac Arrhythmias Based on Electrocardiogram Data Using a Convolutional Neural Network Model

Cardiac arrhythmias are a disease with considerable incidence and prevalence worldwide, and their diagnosis can be complex due to the existence of different types of arrhythmias that share similar characteristics and make an accurate diagnosis difficult. Making a correct diagnosis of the kind of arrhythmia that affects an individual is important to define the most appropriate type of treatment for the case. Machine Learning and Deep Learning techniques have been proposed to automate the diagnosis of arrhythmias to assist healthcare professionals in decision-making. This study proposes a Convolutional Neural Network model for classifying cardiac arrhythmias using electrocardiogram data. The objective is to present a model that achieves high accuracy rates in identifying types of arrhythmias and presents an adequate balance between performance and computational costs. The model was trained with a dataset composed of electrocardiogram exams with 32 types of arrhythmias. In the pre-processing phase, the dataset was restructured to allow the data to be treated as a time series to explore the potential of Convolutional Neural Networks in dealing with data organized in this way. Training was carried out using a state-of-the-art Deep Learning model and the model achieved an accuracy rate of 98.37% in its predictions. This excellent performance confirms the ability of Convolutional Neural Networks to efficiently deal with pattern learning in time series. The results obtained demonstrate the potential of Deep Learning techniques as aiding tools to provide improvements in medical processes.

The Influence of Input Frequency and L2 Proficiency on the Representation of Collocations for Chinese EFL Learners.

Collocations typically refer to habitual word combinations, which not only occur in texts but also constitute an essential component of the mental lexicon. This study focuses on the mental lexicon of Chinese learners of English as a foreign language (EFL), investigating the representation of collocations and the influence of input frequency and L2 proficiency by employing a phrasal decision task. The findings reveal the following: (1) Collocations elicited faster response times and higher accuracy rates than non-collocations. (2) Higher input frequency improved the accuracy of judgments. High-proficiency Chinese EFL learners exhibit better accuracy and faster response times in collocation judgment tests. Additionally, input frequency and L2 proficiency interactively affected both response time and accuracy rate. These results indicate that L2 learners have a processing advantage for collocations, which function as independent entries in the mental lexicon. Both input frequency and L2 proficiency are crucial factors in collocational representation, with increased input frequency and proficiency shifting the representation from analytic retrieval toward holistic recognition in a continuum pattern.

Rapid automatic detection of collapsed buildings with single period LiDAR data after an earthquake

Earthquakes are natural disasters that cannot be determined precisely where and when they will occur. In cases where precautions are insufficient, large losses of life and property can occur. Minimizing the loss of life after an earthquake depends on the rapid identification of collapsed structures and the urgent delivery of rescue teams to heavily damaged structures. Within the scope of the study, damaged and destroyed buildings were automatically detected in a very short time after the earthquake by using the LiDAR point cloud data obtained with a single period flight and the cadastral map of the region. With the algorithm produced, the robust, damaged and destroyed building classes were detected with 98.98% and 98.56% accuracy rates by considering 3D geometric changes in two different study areas. After obtaining LiDAR data, the detection of damaged and collapsed buildings can be performed within two hours. These findings demonstrate the potential of the proposed approach to effectively detect damaged and destroyed buildings after a disaster. The map of damaged and collapsed buildings after the earthquake is produced without being dependent on radiometric changes. This process is carried out in a very short time with a high accuracy rate, which reveals the superiority of the study compared to the literature. The studies conducted with satellite images without 3D analysis capabilities, it is not possible to detect collapsed buildings with undamaged roofs. In contrast, the proposed LiDAR-based method can detect such damaged structures with high accuracy. The method accelerates the planning and implementation of post-earthquake rescue and relief operations. In addition, the fact that it can prevent loss of life due to late interventions increases the importance of the proposed study.

Active noise cancellation using hybrid CNN-LSTM classification and genetic algorithm-based community detection extraction with LMS noise filtering process

Unwanted signals in information-bearing signal referred to as noise could degrade the strength of signals in terms of intelligibility and quality. Over the decade, various researchers developed algorithms to enhance speech signal quality and noise reduction. To address the issue, the study propounded the active noise cancellation method by using a hybridized convolutional neural network–long short-term memory (CNN-LSTM) approach and genetic algorithm (GA)-based community detection feature extraction enhanced with least mean square (LMS) noise filtering process. The quality filtered signals were extracted with feature correlation data and precise relevant features using genetic algorithm-based community detection. The selective parameters aid the classification performance, facilitated by hyperparameter fine-tuning of GA-based community detection. The results of incorporating LSTM will eliminate unnecessary memory content by correlating past information outcomes to classify new feature values. In this method, abnormal and normal signals are classified by our LSTM layers output. These classified outcomes aid in the denoise of active voice signals with a fast convergence rate. The efficiency of the proposed method was assessed in terms of its performance through comparative active noise cancellation (ANC) methods. The higher accuracy rate and low NMSE in the classification of audio signals evidenced the efficacy of the proposed ANC model.

Automating egg damage detection for improved quality control in the food industry using deep learning.

The detection and classification of damage to eggs within the egg industry are of paramount importance for the production of healthy eggs. This study focuses on the automatic identification of cracks and surface damage in chicken eggs using deep learning algorithms. The goal is to enhance egg quality control in the food industry by accurately identifying eggs with physical damage, such as cracks, fractures, or other surface defects, which could compromise their quality. A total of 794 egg images were used in the study, comprising two different classes: damaged and not damaged (intact) eggs. Four different deep learning models based on convolutional neural networks were employed: GoogLeNet, Visual Geometry Group (VGG)-19, MobileNet-v2, and residual network (ResNet)-50. GoogLeNet achieved a classification accuracy of 98.73%, VGG-19 achieved 97.45%, MobileNet-v2 achieved 97.47%, and ResNet-50 achieved 96.84%. According to the results, the GoogLeNet model performed the damage detection with the highest accuracy rate (98.73%). This study encompasses artificial intelligence and deep learning techniques for the automatic detection of egg damage. The early detection of egg damage and accurate interventions highlights the significant importance of using these technologies in the food industry. This approach provides producers with the ability to detect damaged eggs more quickly and accurately, thereby minimizing product losses through timely intervention. Additionally, the use of these technologies offers a more efficient means of classifying and identifying damaged eggs compared to traditional methods.

An active-conductive layer stacked sensor platform for real-time detection of heavy metal ions

The development of electrochemical sensors with high sensitivity and accuracy is crucial for the detection of low-concentration heavy metal ions (HMIs) such as Pb(II) and Hg(II), which remains challenging. To address this issue, MgFe-layered double hydroxide (MgFe-LDH) is synthesized as an active layer on the conductive layer SnO2 loaded with nickel foam (NF) with the assistance of oxygen vacancies (OVs) to form an active-conductive layer stacked sensor. The results show that MgFe-LDH@SnO2/NF exhibits superior sensitivity (Pb(II): 417 μA/μM; Hg(II): 986 μA/μM), wider linear range (1.1–2.2 μM) and lower limits of detection (Pb(II): 1.21 nM; Hg(II): 3.7 nM). The high performance of our sensor is attributed to the improved ability to adsorb HMIs via OV reduced binding energy as well as the promoted rapid electron transport via SnO2 reduced charge transfer resistance, as evidenced by density functional theory and electrochemical measurements. Particularly, by combining the newly proposed tiny electrochemical detection network (ECDNet-tiny) and WeChat mini program, we have built a real-time detection platform for Pb(II) and Hg(II) pollution categories and concentrations, which shows high accuracy (0.98) and recovery rate (91.6 %-106.8 %) in the detection of actual water samples. The qualitative and quantitative calculations of electrochemistry is innovatively implemented by deep learning. Besides, smartphones could display detection results transmitted from cloud servers. Thus, this work provides a new sensor platform to improve the sensitivity, accuracy, and intelligence of detecting quintessential HMIs.

Enhancing Smart Grid Security Using BLS Privacy Blockchain With Siamese Bi‐LSTM for Electricity Theft Detection

ABSTRACTEnergy management inside a blockchain framework developed for smart grids is primarily concerned with improving intrusion detection to protect data privacy. The emphasis is on real‐time detection of cyberattacks and preemptive forecasting of possible risks, especially in the realm of electricity theft within smart grid systems. Existing Electricity Theft Detection techniques for smart grids have obstacles such as class imbalance, which leads to poor generalization, increased complexity due to large EC data aspects, and a high false positive rate in supervised models, resulting in incorrect classification of regular customers as abnormal. To provide security in the smart grid, a novel BLS Privacy Blockchain with Siamese Bi‐LSTM is proposed. Initially, the privacy‐preserving Boneh‐Lynn‐Shacham blockchain technique is built on BLS Short signature and hash algorithms, which mitigate misclassification rates and false positives in the detection of smart grid attacks. Then, a hybrid framework employs an intrusion detection algorithm based on Siamese Bidirectional Long Short‐Term Memory to semantically distinguish between harmful and authentic behaviors, thereby improving data quality and predictive capabilities. Furthermore, a Recurrent Neural Network‐Generative Adversarial Network is presented for detecting electricity fraud, which addresses the issue of class imbalance. This uses both supervised and unsupervised loss functions to produce synthetic theft samples that closely resemble actual theft incidents. From the experiment, it is showing that the proposed models perform with high accuracy and low error rates. The proposed model from the outcomes when compared to other existing models achieves high accuracy, detection rate, recall, and low computation time.