Category Archives: Uncategorized

Authors: Poroni Koiknzi Fousseni, Elvis Thierry Sounna Vofo

Abstract: Long-term dependency modeling remains one of the fundamental challenges in sequence processing tasks across natural language processing, time series analysis, and sequential decision-making. This paper presents a comprehensive analysis of methods for handling long-term dependencies, examining the evolution from traditional recurrent neural networks (RNNs) to modern attention-based architectures. We provide theoretical foundations for the vanishing gradient problem, analyze key architectural innovations including Long Short- Term Memory (LSTM), Gated Recurrent Units (GRU), and Transformer models, and dis- cuss emerging approaches such as State Space Models and Linear Attention mechanisms. Our analysis includes mathematical formulations, computational complexity considerations, and empirical performance comparisons across various sequence modeling tasks. We identify current limitations and propose future research directions for improving long-range sequence modeling capabilities in deep learning systems.

DOI: https://doi.org/10.5281/zenodo.16908505

Authors: Sandipan Chakraborty

Abstract: Back-office logistics and shared services are being re-architected by AI in 2025. Beyond warehouse robots and route optimizers, the largest productivity lift is happening in the “paperwork and pixels” of logistics: order capture, document processing, freight audit & pay, customer service, and compliance. Drawing on current India-market data and public programs (ULIP, GST e-invoicing, ONDC) and global benchmarks (World Bank LPI), this journal synthesizes six concrete AI impacts that leaders can deploy now: (1) intelligent document processing and touchless workflows; (2) predictive ETA and exception control towers; (3) dynamic rating, tendering, and contract optimization; (4) forecasting for capacity, working capital, and SLA staffing; (5) AI copilots for shared-services agents; and (6) digital compliance across GST/e-invoicing and trade. We quantify the opportunity, map enabling Indian rails/APIs, list risks and controls, and close with a practical upskilling and tools roadmap tailored to India..

DOI:

Authors: Piyali Ghosh , Dr. Dhirendra Kumar Tripathi

Abstract: With the accelerated growth of the Internet of Things (IoT), providing secure, scalable, and privacy-preserving communication has become a serious issue. Current routing protocols such as AODV, DSR, and HSRP are incapable of addressing the complex needs of today's IoT systems, particularly in large-scale, heterogeneous, and energy-constrained systems. This paper introduces the Federated Hybrid Secure Routing Protocol (F-HSRP)—a new paradigm that combines federated learning, trust-based routing, AES-256 encryption, and blockchain-aided route verification to address these issues holistically. F-HSRP utilizes a light-weight Convolutional Neural Network (CNN) at the edge of IoT networks. With federated learning, local anomaly detection models are trained at the nodes, maintaining data privacy and facilitating real-time accurate threat detection. Routing decisions are informed through a composite trust score based on node behavior, residual energy, and anomaly scores. Secure data transfer is enabled by AES-256 encryption and a lightweight Proof-of-Authority (PoA) blockchain process that guarantees tamper-proof route verification without imposing substantial overhead. The protocol is tested with the Bot-IoT dataset and a hybrid simulation platform that integrates NS-3 and TensorFlow Federated. The results indicate F-HSRP outperforms conventional protocols with 96.3% anomaly detection rate, 27% energy efficiency improvement, and better packet delivery and delay metrics. It also successfully fights blackhole, replay, and Sybil attacks. By integrating federated intelligence, cryptographic security, and blockchain consensus, F-HSRP offers a strong, energy-efficient, and privacy-enhanced routing solution for real-time IoT applications in smart cities, industrial control, healthcare monitoring, and military systems.