Journal of Computers, Mechanical and Management

Innovative IoT Development

Pandit Darshan Pradeep, Manoj E. Patil — Wed, 30 Apr 2025 00:00:00 +0300

The rapid advancement of the Internet of Things (IoT) is reshaping industries by enabling seamless communication between devices, real-time data collection, and automation. Given the surge in IoT applications, challenges related to security, scalability, and interoperability have become increasingly critical. This study presents a state-of-the-art software engineering framework designed to address these limitations through the integration of blockchain technology and smart contracts. By leveraging the decentralized, immutable, and transparent nature of blockchain, the proposed framework enhances trust and security within IoT environments. Smart contracts, as secure, self-executing code, facilitate autonomous interactions between IoT devices without relying on centralized control. Additionally, the framework introduces novel strategies for optimizing resource management and data handling efficiency while improving system scalability across distributed networks. The synergistic use of blockchain and smart contracts not only resolves key IoT challenges but also contributes to the development of robust, efficient, and scalable IoT ecosystems. The framework is applicable across diverse domains, including healthcare, smart cities, supply chain management, and industrial automation, fostering innovative, self-governing IoT systems throughout their operational lifecycle.

Compare and evaluate AI models for automatically classifying and categorizing URIs

Pooja Tiwari, Ravi Kumar Burman, Abhishek Kumar — Wed, 30 Apr 2025 00:00:00 +0300

URIs, or upper respiratory infections, are among the most prevalent illnesses. Nevertheless, a thorough assessment of the associated burden has not been conducted.Thus, this study's goal is to outline the global and regional burden of URIs. In environments with limited resources, artificial intelligence (AI) systems that use symptoms and signals to identify URTI (upper respiratory tract infection), Pneumonia Bronchiectasis, Bronchiolitis with the help of Such AI systems heterogeneity makes performance analysis necessary to guide future research. Strong evidence exists to encourage more research into machine learning's ability to automatically identify pneumonia based on symptoms and indicators that are easily recognized. Based on the results of this study, suggestions are given for developing and utilizing AI tools, which should enhance the effectiveness of subsequent research.

Blockchain and AI in Fintech: A Dual Approach to Fraud Mitigation

Nikhil Kassetty — Wed, 30 Apr 2025 00:00:00 +0300

This study reviews fraud mitigation in the fintech sector due to the dual application of blockchain and artificial intelligence. Furthermore, the immutability ledger that supports transparency and accountability on the blockchain, and they successfully contain the risk of data and unauthorized financial transactions to some extent. In correlating millions of transactional data items in real-time, applying machine learning algorithms, it detects fraud and predicts it with high reliability. The proposed system describes an AI-based fraud detection system, namely the data acquisition and preparation phase, the training of AI models phase, and the monitoring phase. The advantages of AI-based fraud detection include high accuracy rates, instant monitoring in which a subsequent reaction can be initiated, minimal or no false positives, and cost efficiencies. This survey evaluates and examines current FinTech fraud detection methods, emphasizing the synergistic application of artificial intelligence and blockchain technologies. This study emphasizes implementation based on cost issues, its suitability for expansion and its legal concerns. It is applied to change many businesses, increase efficiency, decrease fraud, and bring innovation to the field of financial technologies.

A Narrative Review of Data Mining Techniques for User Behaviour Recognition with Illustrative Application of the Apriori Algorithm

Sonam, Jyoti — Wed, 30 Apr 2025 00:00:00 +0300

This review examines key data mining algorithms used for user behaviour recognition in computational systems, focusing on frequent pattern mining techniques. We summarize foundational methods such as Apriori, FP-Growth, and ECLAT, comparing their operational principles and limitations. A frequency-based literature analysis shows the widespread use of Apriori in market basket analysis. To illustrate its workings, we include a demonstrative walkthrough of the Apriori algorithm using a hypothetical dataset. The article concludes with insights into performance trade-offs and future directions in algorithmic efficiency.