Browse Items (16481 total)
Sort by:
-
Optimizing Fake News Classification Using Data Fusion and NLP-Based Machine Learning Techniques
In this research, the performance of different machine learning algorithms for identifying fake news using a dataset of news articles labeled as fake or real. The dataset was preprocessed to remove stop words, punctuation, digits, and special characters, and text normalization was applied. Two feature extraction methods, BOW (Bag-of-Words) and TF-IDF, were utilized to convert text data into numerical features. The dataset was split into training and testing phases to train and evaluate models, including Support Vector Classifier, Logistic Regression, Decision Trees, Gradient Boosting Classifier, Random Forest, and Multinomial Naive Bayes. Ensemble models combining various classifiers were also tested. Performance metrics, including precision, recall, and F1-score, were assessed, and confusion matrices were analyzed. Results showed that TF-IDF generally outperformed BOW. The Random Forest model achieved the highest precision (93%) but had a lower recall (83%). The SVC model showed a balanced performance with a precision of 90%, recall of 87%, and an F1-score of 86%. Ensemble models like GB?+?RF exhibited high precision (99%) but lower recall. These findings highlight the strengths of different algorithms in fake news detection and inform the development of practical classification tools. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Optimizing financial fraud detection models using genetic algorithms
In the contemporary financial environment, financial deception is a persistent challenge that results in significant economic losses annually. Using machine learning models to detect fraud has become an essential instrument for financial institutions to mitigate these risks. Nevertheless, the optimization of these models to achieve a balance between efficiency and accuracy continues to be a significant obstacle. In this chapter, the application of Genetic Algorithm (GA) as a potent optimization technique for improving financial fraud detection models is examined. Inspired by natural selection, GAs provide a unique way to addressing complicated optimization problems by iteratively improving a population of solutions. The chapter commences by providing a brief summary of financial detection and the limitations associated with conventional approaches. It then explores the fundamental concepts of GAs, including selection, crossover, mutation, and fitness evaluation, to provide a comprehensive understanding of how GAs may be used to improve fraud detection systems. In an exhaustive methodological section, we explore the actual use of GAs to optimize different model parameters, such as feature selection and hyperparameter tweaking. The analysis shows that GA-optimized models outperform standard approaches in terms of detection accuracy, false-positive rate, and computing efficiency. 2025 selection and editorial matter, Sulabh Bansal, Aprna Tripathi, Shilpa Srivastava and Prem Prakash Vuppuluri; individual chapters, the contributors. -
Optimizing Food Production with a Sustainable Lens: Exploring Blockchain Technology in Raw Plant Materials and Organic Techniques in Achieving Sustainable Development Goals
Amidst a rising population and mounting environ- mental concerns, India seeks a transformative approach to ensure food security and sustainable agriculture by 2030, as outlined in Sustainable Development Goal 2 (SDG 2). This research explores the immense potential of organic farming methods and raw plant materials to unlock this vision. Plants have a wealth of unrealized potential that extends beyond their conventional functions. The study looks at how different plant parts, like branches, leaves, stems, and even "waste"materials, can be used in a variety of ways to increase self-sufficiency, lessen environmental impact, and access renewable resources. Case studies from across the globe highlight this potential, highlighting the many advantages for the environment and communities. Additionally, the study investigates the innovative use of blockchain technology to promote a more transparent and resilient agricultural environment in India. Imagine blockchain-powered climate-smart practices, safe and transparent transactions, and precision agriculture led by sensor data. Water-efficient irrigation, environmentally friendly pest control, and strong traceability systems are all part of this vision, which aims to strengthen the Indian agricultural sector's resilience. The study suggests a framework of customized policy recommendations centered on non-losable farming methods in recognition of the need for wider implementation. This framework, created especially for the Indian context, supports the promotion of agrotourism, improved education and extension services, accessible financial risk management tools, and the smart redistribution of subsidies. The research highlights the transformative potential of this approach by highlighting the many benefits of these practices, including the environmental (less water use, increased biodiversity, improved soil health, and carbon sequestration), social (better community resilience, food security, farmer income, preservation of cultural heritage, equitable trade), and economic (premium market access, lower input costs, and higher yields) gains. In the end, this research offers a strong plan of action for India to greatly advance SDG 2 and create a more sustainable future for all of its people. A food system that feeds people and the environment can be developed by carefully using organic farming methods and unprocessed plant resources in conjunction with successful legislative initiatives. 2024 IEEE. -
Optimizing Fraud Detection Systems in Credit Card Transactions Using Machine Learning Techniques
Rapid e-commerce services and emerging technologies have grown to use credit card usage as a widespread way of effecting payments, thereby increasing bank transaction volume. It is, therefore, equally increasing fraudulent activitiesthus showing the critical need for fraud detection methods development. Class-weighting hyperparameters are studied and applied to handle class imbalance between fraudulent and legitimate transaction classes. We mainly use Bayesian optimization for these hyperparameters tuning with consideration of unbalanced data problems. The key components of our method involve weight-tuning as a preprocessing step and using the extreme gradient boosting [XGBoost] algorithm to enhance further the light gradient boosting machine [LightGBM] based on an ensemble voting process. Moreover, we use deep learning for hyperparameter tuning with special consideration given to our introduced weight-tuning approach. Experiments on real-world datasets demonstrate the efficiency of our strategies. We follow recall-based metrics and the widely used ROC-AUC scores for the unbalanced datasets, which are more appropriate for measuring the model performance. All the algorithms are compared based on fivefold cross-validation, while the majority voting ensemble method is applied to evaluate the combined performance of the algorithms. The previous results prove that LightGBM and XGBoost perform best, with optimal performances obtained at ROC-AUC scores of 0.95, precision of 0.79, recall of 0.80, and an F1 score of 0.79. Further, deep learning with Bayesian Optimization achieves the ROC-AUC scores of 0.908, precision of 0.96, recall of 0.82, F1 score of 0.88, and Accuracy of 0.9996all of which were significant improvements over the previous approaches. This paper presents Bayesian-optimized LightGBM for fraud detection, where it improves accuracy and efficiently tunes hyperparameters. The main novelty here is integrating Bayesian Optimization into dynamically enhancing model performance for handling class imbalance and reducing false detections. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026. -
Optimizing Functional Feed for Growth and Pathogen Resistance in Oreochromis niloticus using Fermented Seaweeds: A Comprehensive Approach Through Solid State Fermentation and Oxidative Stress Response
The study aimed to explore the potential of seaweeds Sargassum wightii and Gracilaria corticate fermented using Bacillus subtilis MN960600 (CK4). Fermented seaweeds showed enhanced antioxidant activity in DPPH assays. A second-order model known as Box-Behnken was used to create an optimized quadratic design for fermentation parameters enhancing protein, reducing sugars, and lipid yields. This optimized feed demonstrated significant growth improvement of 18 to 20 % in Oreochromis niloticus when compared to commercial feed and a 35 to 40% higher growth in fermented and non-fermented feed groups. Additionally, fish fed formulated seaweeds exhibited resilience to Vibrio harveyi and Aeromonas hydrophila pathogen stress. Additionally, the study highlighted the ability of the formulated seaweed in reduction of oxidative stress caused by pathogens Vibrio harveyi and Aeromonas hydrophila in Oreochromis niloticus. The study emphasized the potential use of seaweeds and probiotic bacteria as a sustainable aquafeed. 2025, Egyptian Society for the Development of Fisheries and Human Health. All rights reserved. -
Optimizing Healthcare Analytics: A Zero-Inflated Poisson Approach toPediatric Emergency Room Visits
In various fields, the modeling of count data holds significant importance. The Poisson regression model is a commonly utilized tool for this purpose. However, this model assumes that the data has uniform dispersion, a condition often not met in real-world observations. The nature of overdispersion canvary depending on the specific context. When the overdispersion is primarily dueto an excessive number of zero counts, the Zero-inflated Poisson regressionmodel becomes a more suitable choice for modeling count data. The paper commencesby offering a summary of the theoretical foundations of both Poisson regressionand Zero-inflated Poisson regression. To evaluate their performance, usethe Mean-Squared error (MSE) as a comparative metric. Next, apply these modelsto analyze the frequency of hospital emergency room visits by children between 1018 years of age. The overdispersion of the visit count in our dataset is mostly caused by the excessive occurrence of zero counts. The findings demonstratethat the Zero-inflated Poisson regression model outperforms the standard Poisson regression model in terms of MSE. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Optimizing Healthcare: Enhancing Disease Management with Recommendation Systems
This paper explores a data-driven disease recommendation system for medical professionals based on symptoms. The technology examines symptom patterns to recommend diseases from large datasets by utilizing collaborative filtering and data analytics. To provide individualized disease recommendations based on symptom severity, it goes through data preprocessing and uses techniques like collaborative filtering and cosine similarity. Even if the technology is promising, disease predictions might be strengthened. It seeks to support early disease prediction and offer patients and healthcare professionals individualized guidance. This system demonstrates the potential of technology in healthcare decision-making using a basic Tkinter application. More improvements are anticipated as a result of data-driven approach advancements, which will improve patient care and optimize healthcare procedures. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Optimizing Interpretability in Recommender Systems using a Hybrid Model based on Matrix Factorization and Neural Networks
Recommender systems play a crucial role in the direction of user choices in e-commerce, media, and online services, clearly, there is a trade-off between predictive accuracy and interpretability. In this paper, a new hybrid model that combines Matrix Factorization and a Neural Network framework to maximize the performance of recommendation as well as explainability has been suggested. The model uses Latent factor representation of Matrix Factorization to provide the global user item interactions, and the Neural Network component finds nonlinear interaction and contextual patterns in the data. The hybrid architecture is trained and tested on a Kaggle dataset of 100,000 user-item interactions with several numerical and categorical characteristics. It compares to standalone methods in that the system is more superior with an accuracy of 94.5, F1-score of 0.945, mean absolute error (MAE) of 0.087 and root mean squared error (RMSE) of 0.112. It is proven by computational analysis to have efficient training convergence and low inference latency, allowing real-time recommendations on Google Colab. The proposed solution bridges the gap between performance and transparency since it can be applied and is credible by being predictive and understandable at the same time. The study has implications in intelligent, explainable and scalable recommenders systems in diverse areas of application. 2025 IEEE. -
Optimizing Kidney Ultrasound images through Pre-Processing Filters
Medical image processing and analysis have greatly advanced in the past decade, significantly contributing to the diagnosis of various diseases.However, It is crucial to address the need for effective data management in the medical field due to the significant rise in data generation and storage. It necessitates the exploration of compression methods as a means of achieving efficient data handling. Consideration should be given to image processing approaches to minimize redundancy. Ultrasound imaging has gained importance in recent years, but the presence of artifacts in ultrasound images has complicated diagnoses. An evaluation has been performed to identify appropriate Pre-processing techniques for kidney images before extracting kidney features. Observing the sensitivity and calculating the PSNR and MSE of the filtered image are used to assess the applied methods. The results indicate that the median filter is ideal for image quality enhancement, while the Sobel filter is highly effective in detecting kidney edges. 2023 IEEE. -
Optimizing Machine Learning for Product Category Prediction in Digital Wallet Transactions: A Case Study of Feature-Driven Performance
The Digital Wallet transactions is one of the rapid phenomena in the application of technology. There were various studies which explored to application and sophistication of this digital wallet transactions. Based on the secondary data, the researcher developed a model for classifications using machine learning algorithms in Jupyter notebook (Python IDE). In the current study the performance of the machine learning model for classification is conducted on product categories in digital wallet transaction using many features such as product amount transaction fees cashback and encode categorical variables merchant name product name and payment methods. The test results of the classification model show and oral accuracy of the model at 92% with Precision recall and F1 scores averaging up to 0.92. It is noticeable that some of the features such as gas bill electricity bill showed weaker performance suggesting the need for further engineering and model tuning. This provide the deep understanding on how the transactions related features contribute to predicting the accuracy and highlights the potential for improving classification models for financial technology and its applications. The study also provides future directions and implications for the model refinement focusing on improving miss classification in categories. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Optimizing malachite green adsorption with Co-PTC metal organic framework: Insights into mechanisms and performance
The removal of organic pollutants from aqueous environments has garnered significant attention in environmental science and engineering. Metal-organic frameworks (MOFs) have emerged as promising materials for this purpose due to their intriguing structures, high surface area, and perpetual porosity. In this study, we investigate the adsorption performance of Co-based MOF for the removal of malachite green (MG), a common organic dye pollutant. The MOF, abbreviated as Co-PTC is synthesized via a one-pot green approach using perylene-3,4,9,10-tetracarboxylic dianhydride (PTC) as the ligand at room temperature. Basic to advanced characterization techniques are employed to elucidate the structure and interactions within the MOF. Through a comprehensive analysis, the underlying mechanisms governing the adsorption process are explored, and optimization studies have been carried out. Co-PTC in minute amounts exhibits an adsorption capacity of 79.3 % selectively for MG in 50 min. The kinetics and isotherm models governing the adsorption process are well investigated. 2024 Elsevier B.V. -
Optimizing milk run and use of bin-packing in waste collection problems /
International Journal of Engineering & Technology, Vol.7, Issue 4.10, pp.577-579 -
Optimizing operational cost and delivery of online food delivery apps using high-tech vending machines
Consider the present scenario of placing online food orders while traveling, waiting for them, and struggling to collect them on time. This issue can be addressed by creating and deploying a fully functional high-tech vending machine. With the evolution of technology and the necessity for constant improvement in service quality, customers are thriving for a better customer experience. This article aims to design and methodologically assess the importance of installing vending machines around the most crowded public transportation hubs by dispensing purchased food or beverages online. It focuses on providing a convenient delivery mode for online-ordered food at travel boarding points and public gatherings. Vending machines at these locations gather and distribute food to consumers based on orders from online food delivery apps such as Swiggy, and Zomato, thus optimizing and improving the delivery experience. It focuses on optimizing the operational cost associated with online food delivery platforms and reducing the carbon emissions contributed by multiple deliveries that happen towards the common drop-off points. 2024 Srinesh Thakur, Anvita Electronics, 16-11-762, Vijetha Golden Empire, Hyderabad. -
Optimizing Phishing Email Classification Through Scalable Feature Extraction Using MapReduce
A bag of features (BOF) may be made using either map reduction techniques or a combination of a thesaurus and domain knowledge. This research presents the BOFMR (Bag of Features using MapReduce) and BOFWT (Bag of Features with Weighted Terms) algorithms, a scalable and efficient technique for processing large email datasets and generating feature vectors based on pre-defined characteristics. The outcomes from using both BOFs on identical datasets are compared. The algorithm leverages the parallel processing capabilities of the MapReduce framework to handle the extensive data, ensuring performance and scalability. When creating a bag of words from a training dataset, the BOFMR technique is useful. The map-reduce technique will help to create a bag of features faster even in case of a larger chunk of data. In this experiment, as data size was limited, the performance of map reduce was not measured. In another BOFWT approach, the building of BOF with domain knowledge by using the word thesaurus was a challenge. The experimental result shows that the results of BOFWT are nearer to the output of BOFMR, and both algorithms show the highest accuracy among other methods. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Optimizing Portfolio for Highly Funded Industries Within Budget Constraints for the Period of 20232024
This research paper aims to analyze and optimize portfolios for the top funded industries based on the budget23. The study uses a data-driven approach to identify the best investment opportunities within these industries. The methodology involves collecting financial data, conducting market analysis, and using optimization techniques to create an optimal portfolio. The results of the study show that the top funded industries have a high potential for growth, and the optimized portfolios can maximize returns while minimizing risk. The findings can provide valuable insights for investors and fund managers who are seeking to make informed investment decisions in these industries. The study also highlights the importance of considering the budget constraints while optimizing portfolios. The Author(s), under exclusive license to Springer Nature Switzerland AG 2024. -
Optimizing Resource Allocation in Smart Healthcare Edge Networks Using Federated Swarm Intelligence and Artificial Neural Networks
Smart healthcare edge networks should be able to serve two purposes at once: to train federated machine learning models across a range of devices without violating patient privacy and to schedule other activities with latency constraints, like real-time patient events. Such methods as FFL-ANN attempt this by using fixed fuzzy rules, which do not work in the situation where the conditions of the network change in an unforeseen manner. In this paper, the framework FSI-ANN is introduced to combine particle swarm optimization to quality-aware model aggregation with ant colony optimization to adaptive real-time task scheduling and ANN-based predictions into a single framework. We experimented with FSI-ANN on 200 edge devices. It achieved 0.825 precision compared with 0.82 with FedAvg and 0.80 with FFL-ANN and reduced inference latency by 18%, 0.370.45 s. Throughput was maintained at 33 tasks/sec as compared with 27 of FedAvg. At burst load, the miss rate of the critical deadline was decreased by 90.2 percent and the energy consumed was decreased by 14.8% per round. The results suggest that adaptive learning using swarm is superior to the fixed rule-based approaches and simple averaging in the distribution of resources at the sustainable healthcare advantage. Copyright 2026 K. Praghash et al. International Journal of Distributed Sensor Networks published by John Wiley & Sons Ltd. -
Optimizing resource management using hybrid metaheuristic algorithm for fog layer design in edge computing
The growing complexity of management in fog computing environments necessitates more efficient algorithms capable of optimizing resource allocation, minimizing latency, and maximizing throughput and energy efficiency. Existing techniques, consisting of the Multi-Objective Crow Search Algorithm (MOCSA) and Fuzzy Meta-Heuristics Optimization (FMHO), regularly suffer from suboptimal performance due to constrained exploration abilities and slower convergence fees. To overcome with these demanding situations, this paper proposes a singular Hybrid Metaheuristic Algorithm (HMA) that mixes the strengths of more than one metaheuristic techniques, along with genetic algorithms, simulated annealing, and gray wolf optimization (GA-SA-GWO). The HMA is specifically designed to enhance useful resource control in fog computing by optimizing useful resource allocation, lowering latency, and enhancing usual gadget performance. Experimental results exhibit that the proposed HMA significantly outperforms existing solutions, with 26.98 % improved latency, 90.64 % resource utilization, 96.05 % throughput, 37.06 % reduced energy utilization, and 93.85 % energy utilization. These outcomes spotlight the HMA's potential to successfully manage sources in dynamic and unpredictable fog computing environments, providing a greater scalable and robust solution for actual-time applications. 2025 -
Optimizing Retail Operations with Big Data-Driven Insights: From Inventory Management to Personalized Marketing
In this paper, we look into the role of big data analytics in the strategic transition of retail businesses particularly on inventory management, supply chain and marketing. Using such technologies as big data and machine learning, retailers can find new patterns within such information that can lead to improved efficiency, and satisfaction of consumers. The study also shows noteworthy performance gain in areas of stock out and overstock, inventory Turns and delivery correctness. Even more, the approach to the customer targeting, that stemmed from the principles of the customer segmentation and recommendation systems, led to the growth of the conversion, customer loyalty, and customer lifetime value. The research evidence suggests that information-based management strategies contribute to organizational performance and sustained competitive advantage of firms operating in the retail sector. Issues like data integration, privacy and infrastructural, components are also addressed and hence making it easy form the basis of any future learning and trying out on the real life challenges. The current research focuses on the significance of big data for designing growth and innovation strategies in the changing retail environment. 2025 IEEE. -
Optimizing supercapacitor electrodes via lithium-induced JahnTeller modulation in CuO
AbstractThe development of advanced electrode materials with superior electrochemical properties is essential to meet the growing demand for efficient energy storage technologies. While surface engineering is common to address this fundamental challenge, the present work shifts the focus from external morphology to internal structural stabilization. Through an integrated experimental and density functional theory (DFT) approach, we demonstrate that a moderate lithium incorporation of 4at. % achieves an optimal balance in CuO properties by suppressing subtle JahnTeller distortions, enhancing crystallite size, narrowing the band gap, and improving both optical and electrical conductivity. X-ray Absorption Spectroscopy (XAS) confirms that Li-ion incorporation increases local symmetry around Cu sites, while EXAFS analysis identifies localized structural disorder associated with dopant substitution. This dual effect stabilizes the CuO lattice while simultaneously creating additional redox-active sites. Electrochemical testing validates this approach, as the optimized 4at. % Li-doped CuO electrode delivers a high specific capacitance of 656F/g at 1 A g?1. The fabricated symmetric supercapacitor device delivers an energy density of ~7Whkg?1 at a power density of ~700Wkg?1, demonstrating the feasibility of Li-doped CuO thin films for supercapacitor applications, although further optimization is required to improve long-term cycling stability. This synergistic experimentaltheoretical framework provides both fundamental insight and practical guidelines for the rational design of doped transition-metal oxides, offering a cost-effective and scalable strategy for next-generation energy storage applications. 2026 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies. -
Optimizing Sustainable Agriculture Through Customized Crop Management Approach
Sustainable agriculture is essential for actively addressing the dual challenges of global food security and climate variability. This study deploys an intelligent, data-driven approach to tailored crop management, positioning it as a dynamic framework for optimizing cultivation by enhancing soil quality and climate resilience. Recognizing soil quality as a foundational element of sustainable agriculture, this paper highlights its critical role in nutrient cycling, water retention, and organic enrichment. Through strategic interventions such as precision crop rotation, conservation tillage, cover cropping, and organic amendments, this approach maximizes soil porosity, health, and fertility while mitigating environmental degradation. By operationalization tailored crop management as an adaptable and scalable system, this research advances the synergy between agricultural productivity and environmental sustainability. Leveraging AI-driven insights, predictive modeling, and modular frameworks, these strategies empower global efforts toward food security, ecological balance, and climate-adaptive farming solutions. 2025 IEEE.