Abstract-The rapidly evolving e-commerce landscape demands dynamic pricing strategies to maximize revenue and maintain a competitive edge. This study examines the integration of machine learning (ML) and business intelligence (BI) to enhance pricing strategies, addressing the shortcomings of outdated models in adapting to digital market shifts. While ML has proven valuable in various business applications, its potential for dynamic pricing in e-commerce remains underexplored, particularly when combined with BI. Existing research lacks a comprehensive analysis of how these technologies can work together for pricing optimization. To bridge this gap, the study employs the Support Vector Machine (SVM) algorithm, known for handling complex and nonlinear relationships in large datasets. By leveraging BI tools to collect, process, and analyze crucial data, the approach establishes a real-time pricing framework. The findings reveal that ML-powered BI systems significantly enhance a company’s ability to set accurate prices and swiftly respond to market fluctuations. The adaptability of the SVM model ensures pricing decisions are both precise and responsive to dynamic market conditions, leading to a more effective and competitive pricing strategy.

Abstract-Accurately estimating traffic states is a crucial aspect of transportation engineering, enabling effective traffic control and operations. In recent years, Physics-Regulated Deep Learning (PRDL) has gained significant attention due to its ability to achieve higher accuracy while requiring less training data compared to conventional deep learning (DL) approaches. However, a key challenge of PRDL is the lengthy training time required for closely related but distinct tasks.To address this limitation, this paper introduces a hybrid physics-regulated deep transfer learning approach that leverages the strengths of transfer learning, PRDL, and DL to enhance estimation accuracy and reduce computational costs, particularly in scenarios with limited observation data. The proposed framework includes two transfer learning variants designed to extract and transfer essential features from pre-trained models to new but similar traffic environments. This hybrid approach integrates deep learning training, minimizing computational overhead by eliminating physics-based loss calculations during training.Simulation results demonstrate that, compared to traditional PRDL methods, the proposed transfer learning approaches improve estimation accuracy by over 12% on average while reducing training time by more than 50% on average. These findings highlight the potential of hybrid transfer learning techniques in accelerating the adoption of PRDL for traffic state estimation, making it a valuable tool for transportation systems with limited computational resources.

Abstract-Thyroid disease is one of the most prevalent illnesses worldwide, affecting over 42 million individuals in India alone. The thyroid gland, a small organ located in the neck, plays a crucial role in regulating metabolic processes by secreting essential hormones. Any dysfunction in the thyroid gland can significantly impact overall health. Accurate testing for thyroid disorders is vital for effective treatment, as early diagnosis can help balance hormone secretion and mitigate related complications.However, the increasing number of thyroid patients and the shortage of medical professionals pose challenges to traditional diagnostic methods. To address these issues, a deep learning-based Multi-Layer Recursive Neural Network (ML-RNN) is employed to enhance diagnosis. This approach focuses on preprocessing the input data, selecting relevant features from standard datasets, extracting key attributes, and classifying thyroid conditions into normal, hyperthyroid, and hypothyroid categories.The first stage of this process involves preprocessing, which includes data cleaning, splitting, and handling missing values to enhance data quality. Next, feature selection is performed using the Fisher score method to identify an optimal subset of features. Data analysis is then conducted based on Region-of-Interest (ROI) volumes. Finally, classification is carried out using ML-RNN, which improves accuracy in detecting thyroid disorders and assessing the risk of developing the disease. The model demonstrates high performance in terms of accuracy, recall, positive predictive value, and negative predictive value, making it a reliable tool for thyroid disease prediction.

Abstract-This paper proposes a novel strategy for mitigating catastrophic forgetting of lifelong learning via memory-augmented large language models. Coupling external memory modules with standard deep learning frameworks, our methodology enables the model to retain context information over long periods of time and retrieve such information, preventing previously learned facts from being overwritten by new input data. We demonstrate our approach on the real WikiText-103 dataset, with the results of our experiments showing an extensive improvement in the retention of long-term dependencies and overall model performance. Our findings suggest that memory augmentation is a promising means to enhance the resilience of language models in ever-changing, dynamic settings and laying the groundwork for more robust and adaptable continual learning systems.

Abstract-Heart disease remains one of the leading causes of mortality worldwide. Early detection and prevention are crucial to reducing the risks associated with cardiovascular diseases. This paper presents a rule-based predictive system for heart disease diagnosis. The system leverages patient data, including medical history, lifestyle factors, and clinical symptoms, to provide an accurate assessment of potential heart disease risks. Implemented in Python, the system follows a structured decision-making approach based on medical guidelines and expert knowledge. The proposed system aids in decision-making for healthcare professionals and individuals by evaluating multiple health indicators to predict heart disease risk. This research paper discusses the dataset used, feature selection, rule formulation, system architecture, and evaluation criteria. The study demonstrates that expert-driven approaches can enhance early detection and contribute to improved patient outcomes.

Abstract-Turmeric, a key agricultural product, holds significant economic importance for India, with Erode District in Tamil Nadu being one of the largest turmeric production hubs. However, turmeric exporters from Erode face a series of challenges that impede their competitiveness in the global market. These challenges include inadequate infrastructure, fluctuating market prices, lack of quality standardization, and the complexities of international trade regulations. Additionally, issues such as inconsistent supply due to climatic variations, pest attacks, and post-harvest losses further exacerbate the situation. This study explores these challenges in detail, focusing on their impact on the growth and sustainability of turmeric exports from Erode District, and provides recommendations for overcoming these hurdles through policy interventions, technological upgrades, and market diversification.