Research Briefs on Information and Communication Technology Evolution

Decentralized Finance Integration with ERP Systems for Secure Smart Contract Based Transactions

Nagendra Harish Jamithireddy — 2025-03-11

This study proposes a decentralized framework that merges smart contract based Decentralized Finance (DeFi) protocols and traditional Enterprise Resource Planning (ERP) systems to provide secure, automatic, and verifiable transaction execution. It constructs an additional middleware interface to guarantee interoperability between ERP modules and blockchain networks that utilize smart contracts for procurement, finance, and asset management modules. The system was tested empirically within a hybrid testbed of chains with Ethereum Virtual Machine (EVM) compatibility simulation executing ERP transaction testing on a simulated environment with physical hardware. According to quantitative assessment results, performance increased, achieving a 38% increase in transaction throughput, a 27% decrease in execution costs, increased trust and traceability due to cryptographic audit trails, and improved auditability. The research highlights the potential of DeFi integrated ERP systems for decentralized enterprise finance systems as a scalable secure replacement to centralized enterprise finance systems.

Neural Network Powered Indexing Techniques for High Performance Data Retrieval

Ankita Sappa — 2025-03-11

Modern applications’ data growth poses the limitation of requiring highly effective and flexible indexing methods. Structures such as B-tree, hash index, and spatial tree are considered the oldest forms of indexing, but they seem to fail when it comes to high and constantly changing dimensional structures. The objective of this paper is to increase the speed, precision, and scope of data retrieval enhancement using the indexing methods powered by neural networks. We teach neural networks that cleverly change the key space to storage space as a mapping function to minimize query time and enhance retrieval accuracy. The proposed framework has been tested thoroughly against learned index and classical index baselines on several datasets and workloads. The findings are substantial concerning response times greatly improving with lower memory use and tracking more data changes. Besides filling the gap between deep learning and systems optimization, this paper marks the first step towards building intelligent indexing systems. Deep learning coupled with fast changing data environments challenge the future of indexing systems, and this document is the groundwork for such advanced systems.

Reinforcement Learning Based Optimization of Query Execution Plans in Distributed Databases

Srikanth Reddy Keshireddy — 2025-03-11

Troublesome workloads, data heterogeneity, and shifting resource conditions make efficient query execution highly difficult to achieve in distributed database systems. Traditional optimizers will almost always rely on handcrafted methods or static cost models to achieve the desired results, resulting in adaptative failures along the way and serving at best subpar query execution plans (QEPs). This paper presents a new architecture meant to optimize QEPs by utilizing deep policy reinforcement learning (RL) for dynamically shifting execution strategy adaptations over distributed nodes. The proposed model considers and structures the optimization problem as a Markov Decision Process (MDP) with states available in the form of system and query profiles, actions available being the choices of QEPs, and the rewards acting as a mere performance measurement for execution. We analyze this approach with different combinations of queries and nodes through benchmark datasets and simulated environments. The objective of this evaluation is to test the model’s performance in regards to differing query kinds and node configurations. The experiments indicate remarkable advances in system throughput and execution time while achieving strong generalization to unfamiliar queries. These results support the hypothesized ability of query processing in future distributed databases to not have suggestion mechanisms reliant on rules or costs, unlike their predecessors, and instead implement optimizers that utilize RL.

Automation of SAP ERP Processes Using Agentic Bots and UiPath Framework

Naren Swamy Jamithireddy — 2025-03-11

This research focuses on an automation approach to implementing SAP ERP processes with the help of agentic bots created in the UiPath RPA platform. As enterprise resource planning systems become increasingly complex, manual processes using SAP from the modules becomes more time-consuming, increases errors, and results in highly detailed operational workflows. Using appropriate intelligent, event-driven agentic bots, and combining them with automation functionalities of UiPath, this research exhibits remarkable efficiency improvements in finance, purchasing, and human resource SAP modules. The method used was partitioning of the SAP domain into set SAP tasks that can be automated using reusable automation patterns with the support of custom bots, then measuring how fast the bots could execute the tasks compared to doing them manually. Experiments indicate that the bots were able to perform tasks with up to 72% less execution time, while also displaying more precise results and better handling of exceptions. In addition, the paper looks at scalability, organizational acceptance, and post-deployment results with different business situations. These results highlight the cognitive automation paradigm on enterprise systems as well as suggest a base model for transforming SAP systems based on RPA technology.

Analysis and Classification of COVID-19 Severity Using Machine Learning Techniques

Chandini Krishna Polaki — 2025-07-06

The COVID-19 pandemic has placed significant burdens on the human race over the last few years. In such surveillance, higher data analysis is required with advanced predictive modeling. Herein, classification is performed on the outbreak severity using a publicly available dataset of the daily and accumulated case and mortality data concerning COVID-19. After preprocessing, real-time missing and inconsistent values are replaced to enable feature derivation, including rolling averages. Interactive Tableau dashboards are developed to visualize severity classifications, regional trends, and temporal variations, providing dynamic insights into outbreak patterns. Comparative analysis reveals disparities in case distribution across continents, identifying major hotspots. Machine learning models are employed to predict severity levels, achieving strong performance metrics. The KNN model yields the highest accuracy of classification, which is 97.11%, while the Random Forest model is more resistant to noise and enhances the stability of the predictions. These results emphasize the potential role of machine learning and data visualization examples with Tableau for data-driven public health strategies in monitoring and responding to outbreaks.

Implementation and Evaluation of Evacuation Shelter Support System

Yuta Seri — 2025-07-23

This paper discusses the implementation and evaluation of an evacuation shelter support system. The proposed evacuation shelter support system comprises an evacuee arrival/departure registration management system that utilizes facial recognition technology and an evacuation shelter management system for local government staff. In the proposed system, facial recognition is performed using a camera-equipped tablet device installed at the evacuation shelter to manage the arrival and departure of evacuees. In addition, the proposed system visualizes the congestion status of the evacuation shelters in real time and automatically calculates the required amount of relief supplies from the real-time evacuee information acquired at each evacuation shelter. To evaluate the proposed system, we surveyed 31 participants regarding the system’s usability and effectiveness, and the proposed system received high evaluation scores in many evaluation categories.

A Study on White-Box Cryptography based Integrity Verification Techniques for On-Device AI Model and Their Performance

Han Bin Lee — 2025-12-23

This paper studies White-Box Cryptography (WBC)–based integrity verification for protecting on-device AI models deployed on resource-constrained and potentially hostile platforms. Conventional hash- or MAC-based integrity checks assume that keys and verification logic are isolated from adversaries, an assumption that fails when attackers control the execution environment and can inspect or modify AI binaries and models. WBC embeds secret keys inside heavily obfuscated implementations, making integrity checks more resilient against static and dynamic analysis and key extraction. In this study, methods of using WBC, which offers such security advantages, are investigated to protect on-device AI embedded within embedded systems. To assess the feasibility of WBC-based protection techniques for on-device AI, existing WBC-based integrity verification approaches are selected, implemented, and tested in constrained benchmark environments. Using the algorithmic analysis and experimental results, functional and non-functional requirements are then derived.

A Survey on Data Visualization Techniques for International Trade

Anthony Bias — 2025-12-23

In the globalized world we live in, international trade plays a massive role in providing people with products from bananas to computers. With this in mind, a good understanding of international trade is important for a firm understanding of the world and essential for any large private or state actors to plan policy. However, with the sheer number of different products to trade and countries to trade them, there is a lot of data to analyze. Data visualization can show this data in a much easier form for humans to understand and analyze. This paper explores various techniques and tools for visualizing international trade data after a brief look at some relevant concepts.

Survey on Electric Vehicle Charging Systems: Derivation of Security Requirements and Research Directions

Soowang Lee — 2025-12-26

The proliferation of electric vehicles (EVs) is shifting the paradigms of transportation and energy, and electric vehicle charging infrastructure (EVCS) is emerging as a key element for achieving sustainable mobility. From a systems perspective, EVCS consists of heterogeneous components, including power electronics, communication controllers, and cloud-based management platforms, which are interconnected via standard protocols such as ISO 15118 and OCPP, yet still face technical challenges such as grid instability, lack of interoperability, and high deployment costs. From a security perspective, EVCS functions as a cyber-physical interface between vehicles and the grid and thus presents a broad attack surface, with threats such as charge manipulation attacks (CMA), distributed denial-of-service (DDoS), false data injection (FDI), and man-in-the-middle (MitM) attacks being actively reported. To address these challenges, various AI-based security techniques—including machine learning-based anomaly detection, transfer learning, federated learning, and temporal convolutional network (TCN)-based intrusion detection systems—have been explored; however, their practical deployment remains limited by insufficient datasets, computational resource constraints, and hardware compatibility issues. This paper comprehensively examines EVCS from both systems and security perspectives by analyzing recent charging technology trends, grid integration strategies, standardization issues, and major security vulnerabilities and countermeasures. Building on this analysis, this paper derives security requirements for EVCS and proposes future research directions for realizing more efficient, reliable, and secure charging infrastructures.

Trends in Ransomware Attacks: Infiltration and Encryption Mechanisms of LockBit, Hive, and Akira

Donghwoo Cho — 2025-12-26

In 2024, approximately 65% of financial institutions worldwide reported experiencing ransomware attacks. To respond to these attacks, multiple countries engaged in international collaboration. However, the average ransom-ware payment in 2024 increased fivefold compared to the previous year. This indicates that the ransomware threat remains persistent. In particular, the spread of the RaaS (Ransomware-as-a-Service) model and the sophistication of attack techniques have increased the importance of technical ransomware analysis. This paper summarizes existing technical analysis studies on Lock-Bit, Hive, and Akira ransomware and examines the evolution of ransom-ware, focusing on technical changes. Specifically, we distinguish between in-filtration and encryption methods to derive generalized evolution patterns and discuss directions for future work.