Journal of Science Engineering Technology and Management Science

Drug inventory and supply chain management is a crucial aspect of the pharmaceutical industry that ensures the efficient tracking, storage, and distribution of medicines. this project aims to develop a webbased application using python, django, and sqlite/mysql to streamline the management of drug inventory from manufacturers to pharmacists. the proposed system provides a centralized platform for tracking medicine stock, managing orders, and enhancing supply chain visibility. the platform allows different stakeholders, including administrators, manufacturers, distributors, suppliers, and pharmacists, to efficiently manage their respective roles. by integrating qr code-based tracking and automated order processing, the system improves accuracy, reduces wastage, and prevents stock shortages

Keywords :

Author : K.Nikhitha, S.Raju, Y.Manaswini, B. Kumar, Mrs.G.Archana

Title : AN INTERACTIVE DRUNG SUPPLY CHAIN TRACKING SYSTEM USING BLOCKCHAIN

Volume/Issue : 2025;02(08)

Page No : 402-410

Cloud storage has become an essential service for most users handling large amounts of data. An enhanced cloud storage auditing scheme with deduplication guarantees data integrity and saves storage costs by eliminating redundant copies of the same data while ensuring secure and efficient verification techniques. This project is aimed at designing a system where users can safely store gigabytes of data on cloud servers without worrying that their data is compromised. The deduplication mechanism also ensures that a single copy of duplicate data is stored, greatly conserving storage space and minimizing costs. The system also facilitates dynamic operations like data updating, deletion, and modification while upholding high security. To avoid unauthorized access during deduplication, the scheme utilizes encryption methods that permit only authorized users to access and authenticate their data. This method improves data confidentiality, integrity, and availability in cloud systems. The project also maintains low computational overhead for both cloud servers and users, hence a viable solution for contemporary cloud storage requirements

Keywords :

Author : N.Sai Srujana, M.Anusha, M.Manoj, Mrs. O. Naga Kumari

Title : IMPROVED CLOUD STORAGE AUDITING SCHEME USING DEDUPLICATION

Volume/Issue : 2025;02(08)

Page No : 411-419

This paper presents a review kind of data compression ways. Data compression is extensively used by the community because through a compression we can save storehouse. Data compression can also speed up a transmission of data from one person to another. In performing a compression requires a system of data compression that can be used, the system can also be used to compress a data. Data that can be compressed not only textbook data but can be images and videotape. Data compression fashion is divided into 2 videlicet lossy compression and lossless compression. But which is frequently used to perform a compression that's lossless compression. A kind of lossless condensing similar as Huffman, Shannon Fano, Tunstall, Lempel Ziv welch and run- length encoding. Each system has the capability to perform a different compression. This paper explains how a system works in doing a compression and explains which system is well used in doing a data compression in the form of textbook. The affair generated in doing a can be known through the compression train size that becomes lower than the original train.

Keywords :

Author : M. Sujith, T. Bharath Reddy, P. Ashok, T. Sidhartha Reddy, Mr. R. Bharath

Title : DATA COMPRESSION FOR BACKBONE NETWORK SECURITY

Volume/Issue : 2025;02(08)

Page No : 420-427

Low- light image improvement is a grueling task that has attracted considerable attention. filmland taken in low- light conditions frequently have bad visual quality. To address the problem, we regard the low- light improvement as a residual literacy problem that's to estimate the residual between low- and normal- light images. In this paper, we propose a new Deep Lightening Network (DLN) that benefits from the recent development of Convolutional Neural Networks (CNNs). The proposed DLN consists of several Lightening Back protuberance (LBP) blocks. The LBPs perform lightening and darkening processes iteratively to learn the residual for normal- light estimations. To effectively use the original and global features, we also propose a point Aggregation (FA) block that adaptively fuses the results of different LBPs. We estimate the proposed system on different datasets. Numerical results show that our proposed DLN approach outperforms other styles under both objective and private criteria

Keywords :

Author : N. Yashwanth Kumar, P. Veerendhar, Dk. Sai Kumar Reddy, Mr. Kundan. B

Title : A COMPARATIVE STUDY ON CNN-BASED LOW-LIGHT IMAGE ENHANCEMENT

Volume/Issue : 2025;02(08)

Page No : 428-436

The quick headway in Fake Insights (AI) has opened modern conceivable outcomes for upgrading availability for outwardly impeded clients. This venture presents a multimodal GPT-based application that consistently coordinating Normal Dialect Preparing (NLP) and picture acknowledgment innovations to supply comprehensive talked depictions of visual substance. The framework empowers real-time interaction by translating objects, scenes, and feelings in pictures and changing over them into important stories, in this way bridging the crevice between visual and non-visual mediums. The application offers transformative benefits in different spaces. For outwardly disabled clients, it moves forward every day availability by deciphering complex visuals into expressive sound. In instruction, it helps by clarifying classroom visuals, making learning comprehensive. In e-commerce, it encourages nitty gritty item depictions, improving the shopping involvement. Whereas the framework illustrates solid execution with clear visual substance, challenges continue in precisely deciphering complex scenes, enthusiastic prompts, and vague settings, as well as in accomplishing real-time preparing on low- resource gadgets. This extend emphasizes a adaptable and user-friendly plan, utilizing progressed profound learning models such as Convolutional Neural Systems (CNNs) for picture handling and strong NLP strategies for conversational coherence. The design is outlined to handle multimodal inputs successfully, with future cycles arranged to address current restrictions, counting way better scene translation and compliance with information protection directions. By progressing the capabilities of AI-driven conversational interfacing, this extend lays the foundation for more comprehensive innovation arrangements. The continuous refinement and extension of these frameworks have the potential to revolutionize availability, instruction, and client encounters over assorted divisions.

Keywords :

Author : N. Joshika, S. Srikanth, S. Suraj, S. Pradeep, Mrs. A. Nandini Sree

Title : A MULTIMODEL GPT BASED APPLICATION TO HELP BLIND USERS VISUAL PICTURE EXTENSIONS

Volume/Issue : 2025;02(08)

Page No : 437-443

This project introduces an automated method for stitching panoramic images by leveraging machine learning and computer vision technologies. Traditional methods often require manually sorted images, which can be time-consuming and error-prone. The proposed system eliminates this dependency by automatically identifying and grouping related images based on visual similarity. Feature detection is performed using the Scale- Invariant Feature Transform (SIFT) algorithm, which extracts robust keypoints unaffected by scale, rotation, or lighting variations. These keypoints are matched between image pairs using nearest-neighbor techniques, and the Random Sample Consensus (RANSAC) algorithm is applied to compute accurate homography matrices while filtering out outliers.Once the transformation matrices are obtained, the images are geometrically aligned and warped into a common coordinate space to form a cohesive panorama. Image blending techniques such as multi-band blending or feathering are employed to ensure smooth transitions across image boundaries, minimizing visible seams and exposure differences. The system also integrates quality assessment metrics to evaluate the alignment accuracy and overall visual coherence of the stitched panorama. To enhance scalability, the approach supports batch processing of large image datasets and can adapt to unordered or variably illuminated inputs. Potential applications of this system include automated mapping, virtual tours, surveillance systems, and content generation in augmented reality environments.

Keywords :

Author : B. Ajay, G. Shriya Reddy, S. Sai Nishith, Mr. S. Sanjeeva Rao

Title : ROBUST ALIGNMENT FOR PANORAMIC IMAGE STITCHING VIA AN EXACT RANK CONSTRAINT

Volume/Issue : 2025;02(08)

Page No : 444-451

Data grows at the impressive rate of 50% per year, and 75% of the digital world is a copy Although keeping multiple copies of data is necessary to guarantee their availability and long-term durability, in many situations the amount of data redundancy is immoderate. By keeping a single copy of repeated data, data deduplication is considered as one of the most promising solutions to reduce the storage costs, and improve users experience by saving network bandwidth and reducing backup time. However, this solution must now solve many security issues to be completely satisfying. In this paper we target the attacks from malicious clients that are based on the manipulation of data identifiers and those based on backup time and network traffic observation. We present a deduplication scheme mixing an intra and an inter-user deduplication to build a storage system that is secure against the type of attacks by controlling the correspondence between files and their identifiers and making the inter- user deduplication unnoticeable to clients using deduplication proxies. Our method provides global storage space savings, per-client bandwidth network savings between clients and deduplication proxies, and global network bandwidth savings between deduplication proxies and the storage server. The evaluation of our solution compared to a classic system shows that the overhead introduced by our scheme is mostly due to data encryption which is necessary to ensure confidentiality

Keywords :

Author : CH. Deepak Sumanth, G. Manideep, U. Abhinav, B. Vinay Reddy, Mr. Kundan. B

Title : IMPROVED CLOUD STORAGE AUDITING SCHEME WITH DEDUPLICATION

Volume/Issue : 2025;02(08)

Page No : 452-457

Employee motivation plays a critical role in determining organizational success, especially in the Information Technology (IT) sector where innovation, productivity, and adaptability are key drivers. Motivated employees are more likely to demonstrate higher efficiency, creativity, and commitment, leading to improved job performance and overall organizational growth. In the IT industry, where tasks are dynamic and demanding, motivation acts as a catalyst to reduce employee turnover, enhance job satisfaction, and sustain long-term competitiveness. This study focuses on understanding the relationship between motivation and job performance, highlighting both intrinsic factors such as recognition, skill development, and career growth, as well as extrinsic factors like salary, incentives, and work environment. Findings suggest that organizations that implement structured motivational strategies are better positioned to achieve employee engagement, innovation, and sustainable performance outcomes

Keywords :

Author : Niligonda Harish, Srilekha Rageru, M.Pavani

Title : EMPLOYEES MOTIVATION AND ITS IMPACT ON JOB IT

Volume/Issue : 2025;02(08)

Page No : 458-465

The study is undergone at SUNDARAM INDUSTRIES PRIVATE LIMITED ROYAPETTAH, CHENNAI The project focuses on the employee’s job satisfaction level. Job satisfaction refers to one's feelings towards one's job. If the employee’s expectations are fulfilled (or) the employees get higher than what he/she feels with with satisfied. If job satisfaction increases organization commitment will increase This results in higher productivity. The main objectives of this paper are to assess the job satisfaction: to identify the effectiveness of job satisfaction and find out the several factors like personal and organizational factors influencing job satisfaction: to identify the effectiveness of job satisfaction and to find out the several factors like personal and organizational factors influencing job satisfaction of employees. In this study, 95 no's of respondents were been taken as a sample. Percentage analysis and chi-square have been incorporated for research analysis. The study helped in revealing the level of satisfaction of employees concerning the various factors provided in the organization. This study clearly shows that employees under organization are more or less satisfied with the job. The organization should consider the salary, relationship of employees and supervisors, grievance handling, and give more opportunities to the new employees

Keywords :

Author : S.Nithin, V.S.N Reddy, M.Pavani

Title : JOB SATISFACTION AMONG THE EMPLOYEE IN THE SUNDARAM INDUSTRIES PRIVATE LIMITE

Volume/Issue : 2025;02(08)

Page No : 476-485

Recruitment and selection are the most vital functions of human resource management as they directly influence organizational performance and employee satisfaction. In the context of Buzibrains, Bangalore, the recruitment and selection process plays a crucial role in attracting skilled professionals, assessing their potential, and ensuring the right fit between the candidate and the organizational culture. The process not only helps in fulfilling the staffing requirements but also supports strategic business goals by building a competent workforce. An effective recruitment and selection system enhances productivity, reduces employee turnover, and fosters organizational growth. This study focuses on analyzing the methods adopted by Buzibrains for recruitment and selection, highlighting their effectiveness, challenges, and impact on overall business performance.

Keywords :

Author : Thumma Ruthu, M.Ramakanth Reddy, R.Srilekha

Title : RECRUITMENT AND SELECTION PROCESS IN BUZIBRAINS ,BANGALORE

Volume/Issue : 2025;02(08)

Page No : 486-495

The profit that an arranging's workforce determine is densely dependent on the facts frugality of contemporary. Organizations have started repaying excellent consideration to the conscription and election process because public are their priceless property in aforementioned a menacingly aggressive environment. The essential parts of the stick collection process, nevertheless, are without care checked. Prior research on the improve and election process was generally concentrated on judging representative efficiency and the election tests that would take back best choice society, therefore advancing clerk memory and organizational influence. The singular facet concerning this research is that it analyzes the current conscription and choice processes secondhand for one two-fold and after second instruction areas in Australia's city and country extents. The objective concerning this study search out act an practical hearing to define critical items of the enlisting process that can influence selections established miscellaneous shareholder outlooks, containing those of recruiters, favorable aspirants, and failing contenders. To get intuitiveness into by means of what to plan realistic advice for process bettering, any of facets, containing response supplying, interview partnership and readiness, relevance of interview questions, event, and bias, were checked. Their equivalences were too judged.

Keywords : Recruitment, Selection, Challenge, Process

Author : V.Swapna Rani, R.Srilekha, M.Pavani

Title : CHALLENGES FACED IN RECRUITMENT

Volume/Issue : 2025;02(08)

Page No : 496-504

Employee Motivation is the key factor to help up the capacity of an association. In the globalization stage, each organization needs to continue itself in a furiously focused market. Representatives are the fundamental elements to show the business objectives into the real world. Subsequently, in the present world each association attempts to deal with its human asset office to keep its representatives inspired. In that specific circumstance, some of the administration speculations have been rehearsing by them. Business work or their presentation in the market can be assessed by surveying the degree of inspiration of representatives. Inspiration can assume a lead job to get the expert achievement in each budgetary year in a less exertion way. While trying to find what inspires representatives, representative inspiration inquire about papers report that what spurs laborers today is fundamentally connected to the qualities and objectives of the person. Starting in the 1970's, it appears that there was a sensational move in the qualities and needs of the ordinary laborer. As individuals turned out to be progressively taught and the economy saw a sharp decay of the Protestant hard working attitudes, what representatives needed from their managers changed also.

Keywords :

Author : G.Pavani, Ragiri Manisha, R.Srilekha

Title : A STUDY ON EMPLOYEE MOTIVATION ON HDFC

Volume/Issue : 2025;02(08)

Page No : 505-513

Social media marketing is a new trend and the most successful among all the types of marketing tools. For a business to succeed, Advertisement should reach maximum users and create a full clear and visible image of its brand. No doubt, social media marketing is the most cost effective advertising strategy. Most social networking platforms provide free accounts and signup and some online advertising tools come with some cost. On the other hand, the return is much more than investment, in case of social media advertising. Only by investing a little money and time, any one can greatly maximise and increase conversion rate and get high return on investment for the money invested in advertising.The main aim of this study is to examine how social media marketing will affect the final decisions and behaviour of the consumer who spend their maximum time on social networking platforms and forecast relationships among various consumer activities, marketing activities and choice with the behaviour of the consumer.Social media marketing is emerging continuously to satisfy the basic needs of website users and on the other side it also has increased the opportunities for companies to advertise their products and services in personalised way. The previous record shows that social media has contributed greatly in changing consumer behavior and organisations got to know about the customer behaviour.Social media marketing has changed the marketing structure of the place where the goods are sold and purchased online. Social media marketing has significantly changed the way of thinking of the consumer and the organisations should not ignore the social media marketing Growing popularity of social media compelled the marketers to think about this media along with traditional functional areas of marketing. Social media is based primarily on the internet or cellular phonebased applications and tools to share information among people. The number of social media user’s is more than the population of some of the countries today. Impact of social media on marketing can be judged by comparing marketing before social media and marketing after the introduction of social media and the type of technologies used in social media

Keywords :

Author : K.Sai kiran, M.Rajeshwar Reddy, R.Gowtham

Title : A STUDY ON IMPACT OF SOCIAL MEDIA IN MARKETING ON FACE BOOK

Volume/Issue : 2025;02(08)

Page No : 514-524

INTERNET became more powerful and basic tools for every person's need and the way people work by integrating various online management tools using internet, innovative companies have set up the payment system for taking customer orders, facilitate making of payments, customer service, collection of marketing data, and online feedback respectively. These activities have collectively known as e-commerce or Internet commerce. Online shopping made so easy for everyone with their product variations and simple way to buy things. An attempt has been made to critically examine various corporate and business level strategies of two big e-tailers and those are Flipkart and Amazon. Comparison have been done considering e-commerce challenges, their business model, funding, revenue generation, growth, survival strategies, Shoppers? online shopping experience, value added differentiation, and product offerings. Both these big players made their own mark in India, but who is going to be ultimate winner or be the top one is going to be. A comparative study of Flipkart.com with one of the close competitor Amazon.com delivers the information about the different strategies to succeed in e-commerce market and different opportunities available in India.

Keywords :

Author : Makne Ravi, V.S.N Reddy, M.Pavani

Title : COMPARATIVE STUDY AND MARKETING STRATEGIES OF FLIPKART AND AMAZON

Volume/Issue : 2025;02(08)

Page No : 525-533

Network Marketing, also known as Multi-Level Marketing (MLM), is an emerging business model that emphasizes direct selling of products and services through independent distributors. Unlike traditional business structures, network marketing eliminates intermediaries and allows individuals to earn income through personal sales as well as team-based sales performance. The study focuses on understanding the needs, scope, benefits, and limitations of network marketing in the modern economy. It highlights how network marketing has become a popular platform for self-employment, entrepreneurship, financial independence, and personal skill development. With its low investment requirements, flexible work structure, and global reach, it offers a viable alternative to traditional employment. At the same time, the research also considers the challenges and risks associated with this industry, such as misconceptions, market saturation, legal issues, and income instability. Despite these limitations, network marketing continues to grow as a significant contributor to economic development, especially in sectors like healthcare, wellness, cosmetics, and financial services. This study provides insights into the business model, opportunities, challenges, and future potential of network marketing, making it relevant for students, entrepreneurs, and professionals who wish to explore this dynamic field

Keywords :

Author : R.Ajith kumar, CH.Dharmender, M.Pavani

Title : STUDY ON NETWORK MARKETING

Volume/Issue : 2025;02(08)

Page No : 534-540

Work-life balance is a critical issue in modern organizational behavior and human resource management. It refers to the ability of employees to effectively manage professional responsibilities while maintaining personal commitments, health, and social relationships. The rapid growth of technology, globalization, and competitive work environments has blurred the boundaries between work and personal life, making balance increasingly difficult. Lack of balance can lead to stress, burnout, decreased productivity, and higher attrition, while a supportive work-life culture fosters job satisfaction, employee retention, and organizational growth. In the IT and service sectors, where workloads and long hours are common, achieving work-life balance is especially important. Companies that prioritize flexible schedules, remote work options, employee wellness programs, and family-friendly policies demonstrate higher levels of engagement, loyalty, and performance. Thus, work-life balance is not only a personal necessity but also a strategic factor for sustainable organizational success.

Keywords :

Author : P.Sai raghu, K.Shashidhar, R.Manisha

Title : WORK LIFE BALANCE

Volume/Issue : 2025;02(08)

Page No : 466-475

Cloud computing has transformed how businesses and individuals store and access data, but it has also introduced critical vulnerabilities—particularly insider threats. These threats arise when employees or privileged users misuse their access to sensitive information. According to the 2022 Cloud Security Report, insider attacks account for approximately 35% of all global cloud data breaches. Detecting such threats in cloud environments is vital to maintaining data integrity, confidentiality, and business continuity.Traditional detection methods—such as rule-based systems, manual audits, and access log analyses—are reactive, time-consuming, and prone to human error. These conventional systems are often ineffective in large-scale cloud infrastructures, where the volume of data is immense and rapidly changing, making timely threat detection difficult.Given the rising frequency of insider threat incidents and the limitations of existing methods, there is a growing need for more advanced and automated solutions. Machine learning, especially ensemble learning models, provides a promising approach to enhancing detection capabilities. By leveraging algorithms like Random Forest, AdaBoost, and CatBoost, these models can efficiently analyze user behavior patterns, identify anomalies, and detect potential threats in real-time. Unlike traditional systems, machine learning models can process vast datasets from cloud logs and user activities more accurately, reducing false positives and significantly improving the effectiveness of security responses.

Keywords : Cloud security, insider threats, ensemble learning, machine learning, anomaly detection, cloud logs, privileged access, real-time detection

Author : P. Manjulatha, Y. Pavan Sai, G. Rakshit Kumar, K. Chandu Vara Prasad, M. Shiva Prasad

Title : ML-BASED DETECTION AND PREVENTION OF PRIVILEGE ESCALATION ATTACKS IN CLOUD ENVIRONMENTS

Volume/Issue : 2025;02(08)

Page No : 541-547

This research presents an automated news article classification system capable of categorizing content into five categories: business, entertainment, politics, sports, and technology. Manual classification of news articles is often time-consuming, inconsistent, and unable to scale with the rapid growth of digital content. Human annotators may introduce subjective biases, leading to errors and inefficiencies in large datasets. These limitations make manual methods unsuitable for real-time news processing in dynamic environments.To overcome these challenges, the proposed system features a streamlined pipeline integrated into a user-friendly graphical interface. This interface enables users to upload datasets, preprocess the text through cleaning and tokenization, and perform feature extraction using a hybrid Global Vectors for Word Representation (GloVe) approach. This hybrid model combines global word embeddings with context-aware semantic features. The classification task is handled using the Gaussian Naive Bayes (GNB) algorithm.Performance evaluation was conducted by comparing three models. The proposed Hybrid GloVe with GNB model achieved the highest performance, with an accuracy of 91.24%, precision of 90.31%, recall of 92.16%, and an F1-score of 93.19%. In contrast, the existing PCA with GNB model reached an accuracy of 79.10%, and the LDA with GNB model lagged behind with an accuracy of 67.64%. The results demonstrate the superiority of the proposed method in terms of accuracy, processing speed, and consistency, making it a scalable and reliable solution for real-time news classification

Keywords : News classification, GloVe embeddings, Gaussian Naive Bayes, real-time processing, text preprocessing, feature extraction, NLP, PCA, LDA, automated categorization

Author : K. Swayam Prabha, B. B. Praneeth, E. Jaya Prakash, Md. Khalid Hasan, L. Rohith

Title : A Scalable Hybrid GloVe-Based System for Real-Time News Article Classification

Volume/Issue : 2025;02(08)

Page No : 548-555

In India, infrastructure plays a pivotal role in economic development, yet building collapses have increased due to structural weaknesses. According to the National Crime Records Bureau, over 1,200 building collapses were reported annually between 2015 and 2020. Traditionally, civil engineers rely on manual formulas, which lack precision The objective is to predict the compressive strength of buildings Before machine learning, building strength was predicted using empirical formulas, standard material codes (e.g., IS 456:2000), and engineering judgment based on concrete mix ratios, curing time, and load tests. These methods involved trial-and-error concrete mix design, compressive strength tests (e.g., cube testing), and regression through manually plotted graphs. Traditional systems are time-consuming, cost-intensive, and less adaptable to varying construction scenarios. They lack predictive capabilities, struggle with large datasets, and ignore hidden patterns between influencing variables. These drawbacks lead to inaccurate assessments, delayed results, and increased risks in structural safety decisions. This research aims to overcome traditional limitations by introducing machine learning for predictive accuracy. ML models like XGBoost can handle non-linear data, detect complex correlations, reduce human error, and generate quick, real-time predictions. These improvements significantly enhance safety, cost-efficiency, and performance monitoring in modern building projects. The proposed system leverages machine learning models to predict the compressive strength of building materials based on historical data inputs such as cement, slag, ash, water, superplasticizer, coarse/fine aggregates, and curing age. By applying XGBoost Regressor and Random Forest Regressor, the system learns from existing datasets to predict outcomes with high accuracy. Compared to traditional methods, this automated approach provides faster results, handles large datasets effectively, and reduces manual intervention. XGBoost, in particular, yields superior performance making it ideal for real-world deployment in construction sites, improving quality assurance, and aiding in smart civil engineering practices

Keywords : XGBoost, Random Forest, civil engineering, construction safety, predictive modeling, data-driven approach, building materials, infrastructure, India, quality assurance, smart construction, structural integrity

Author : G. Vidyulatha, S. Ajay, V. Snehith Rao, Sk. Arif, K. Sai Krishna

Title : MACHINE LEARNING APPROACHES FOR HIGH-PRECISION CONCRETE STRENGTH PREDICTION IN INDUSTRIAL CEMENT PRODUCTION

Volume/Issue : 2025;02(08)

Page No : 556-566

In today’s digital era, where sensitive data is frequently transmitted over email, ensuring secure communication has become a pressing concern. This paper aims to integrate the principles of quantum cryptography to redefine email security. Quantum Key Distribution (QKD) leverages the laws of quantum mechanics to create cryptographic keys that are virtually unbreakable, offering an unprecedented level of protection against modern cyber threats. Historically, email security evolved alongside the rise of the internet. Initial systems lacked robust protection, relying on basic passwords or unencrypted messages. The introduction of Secure/Multipurpose Internet Mail Extensions (S/MIME) and Pretty Good Privacy (PGP) marked significant advancements, providing encryption for email content. However, traditional encryption techniques depend heavily on the computational complexity of algorithms, making them vulnerable to advances in processing power, particularly with the advent of quantum computing. Traditional email security systems have several limitations. While methods like TLS and end-to-end encryption mitigate some risks, they cannot guarantee long-term security against quantum-level decryption techniques. The motivation for this paper stems from the need to address these vulnerabilities and enhance trust in digital communication systems. Increasingly sophisticated cyberattacks, data breaches, and the potential threats posed by quantum computing have inspired the development of a revolutionary solution that combines quantum cryptography with email systems. The goal is to proactively safeguard sensitive communication against evolving threats, ensuring data integrity and confidentiality. The proposed system integrates QKD to secure email communications. By generating encryption keys based on quantum states, it ensures that any attempt to intercept the keys is detectable. This system also encrypts email content and attachments with these quantum-generated keys, providing enhanced security against interception and decryption attempts. With features like quantum-encrypted file storage and robust authentication mechanisms, this system sets a new standard for email security

Keywords : Quantum key distribution (QKD), email security, quantum cryptography, post-quantum encryption, secure communication

Author : M. Venkatesh, K. Varsha, D. Dileep Kumar, K. Likith Teja, Adnan UL Hassan

Title : QUANTUM-ENHANCED EMAIL SECURITY: LEVERAGING QUANTUM KEY DISTRIBUTION FOR NEXT-GENERATION DATA PROTECTION

Volume/Issue : 2025;02(08)

Page No : 567-577

In industrial robotics, studies indicate that over 30% of unplanned downtime is caused by equipment failures, with robot-related faults accounting for nearly 20% of total annual maintenance costs. Furthermore, predictive maintenance powered by artificial intelligence (AI) has the potential to reduce repair expenses by up to 25% and improve uptime by 10–20%. These statistics highlight the urgent need for intelligent fault diagnosis systems to enhance reliability and efficiency in robotic operations. Traditional manual diagnostic methods are time-consuming, reliant on skilled personnel, and often incapable of detecting early-stage failures in dynamic industrial environments. They also lack the consistency and adaptability required for real-time, sensor-intensive robotic processes, resulting in costly production delays and slow maintenance responses. To overcome these limitations, this study proposes a robust AI-based fault diagnosis system that utilizes sensor data—including force, torque, voltage, and current—from robotic arms. The dataset undergoes comprehensive preprocessing, including outlier removal, normalization, and division into training, validation, and testing subsets. Two machine learning models are implemented: an existing K-Nearest Neighbors (KNN) classifier and a proposed Deep Neural Network (DNN), trained to classify system conditions as either normal or indicative of failure. The DNN architecture, composed of multiple hidden layers, effectively captures complex patterns in the sensor data, enabling accurate fault classification. Model performance is assessed using key evaluation metrics such as accuracy, precision, recall, and F1-score. Once trained, the system facilitates real-time fault prediction and failure pattern analysis, supporting preventive maintenance and significantly improving the safety, reliability, and productivity of industrial robotic systems

Keywords : Industrial robotics, fault diagnosis, predictive maintenance, machine learning, DNN, KNN, sensor data analysis, real-time failure detection, robotic fault classification

Author : Dr Nazimunnisa, C. Karthik, K. Srija Reddy, Kakkula. Arya Kumar Sagar, Sharath Chandra

Title : AI-DRIVEN PREDICTIVE MAINTENANCE FOR ROBOTIC SYSTEMS IN INDUSTRIAL ENVIRONMENTS

Volume/Issue : 2025;02(08)

Page No : 578-589

Dissolved oxygen (DO) plays a critical role in maintaining the health and stability of river ecosystems, as aquatic organisms rely on it for respiration and survival. Therefore, monitoring and accurately predicting DO levels is essential for effective environmental management, conservation efforts, and water quality assessment. Traditionally, DO prediction involves the physical collection of water samples from various river locations, followed by laboratory analysis. Based on the collected data, researchers build empirical models to estimate DO levels using related physical and chemical factors. However, this conventional approach has significant limitations, including data gaps due to spatial and temporal constraints, delayed results, high costs, and limited coverage, which hinder realtime monitoring and accurate forecasting. Recognizing these challenges and the importance of maintaining appropriate DO levels, there is a growing need for more efficient and precise prediction methods. Machine learning (ML) emerges as a powerful solution, capable of analyzing large volumes of data and uncovering complex relationships that traditional models may overlook. This study aims to develop a machine learning-based predictive model for DO levels in river water, leveraging both historical and real-time environmental data. Such a model would enable continuous, reliable DO predictions, supporting better ecosystem health assessment and ensuring the well-being of aquatic life. Ultimately, the findings of this research will contribute to improved environmental management, informed conservation strategies, and sustainable water resource planning for future generations.

Keywords : Dissolved Oxygen Prediction, River Water Quality, Machine Learning, Environmental Monitoring, Aquatic Ecosystems, Real-Time Analysis, Water Resource Management, Conservation, Sustainable Development

Author : P. Anupama, E. Vinay Kumar, S. Ajanth, Kondaveni Anugna, Jeripothula Venkatesh

Title : ML BASED PREDICTING OF DISSOLVED OXYGEN LEVELS FOR SUSTAINABLE RIVER ECOSYSTEM MANAGEMENT

Volume/Issue : 2025;02(08)

Page No : 590-601

In modern healthcare systems, the centralized storage of electronic health records (EHRs) and associated documents presents significant vulnerabilities, including single points of failure, data tampering, and privacy breaches. This research proposes a decentralized framework that integrates blockchain technology with the InterPlanetary File System (IPFS) to securely manage doctor–patient appointment data, medical reports, prescriptions, and feedback. The system utilizes a permissioned Ethereum-style smart contract, named “Healthcare,” to immutably store user profiles, appointment metadata, prescription references, and patient ratings. Sensitive medical files are encrypted using AES-CTR mode, with encryption keys derived via PBKDF2, and pinned to a local IPFS node to ensure both confidentiality and decentralized availability. Patients can register and log in, browse a list of doctors with on-chain average ratings, and book appointments by uploading encrypted medical reports. Later, they can securely download prescriptions in decrypted form. Doctors access the platform to view pending appointments, retrieve patient reports, issue encrypted prescriptions, and publish them on-chain. Patient feedback is also recorded as immutable blockchain transactions. To improve system responsiveness, an in-memory Python cache is synchronized with the blockchain, enabling fast read operations without repeatedly querying the chain. Experimental results demonstrate that the encryption, IPFS pinning, and decryption processes operate within acceptable performance thresholds. The proposed system enhances data integrity, transparency, and fault tolerance: no single authority can alter records without detection, and encrypted files remain retrievable even if specific nodes fail. By combining decentralized metadata storage with secure, off-chain encrypted file handling, this framework offers a robust, patient-centric solution to modern EHR management challenges.

Keywords :

Author : Ch. Aruna, T. Varsha, C. Shivakumar, Mankala Sreeja, B. Badri Nath

Title : A BLOCKCHAIN ENABLED TAMPER PROOF APPROACH FOR ENHANCED MEDICAL DATA INTEGRITY AND ACCESS CONTROL

Volume/Issue : 2025;02(08)

Page No : 602-612

In India, data breaches and unauthorized access to sensitive files have surged, with over 700 million digital records compromised in the last decade. Manual file-sharing methods remain prevalent in sectors like healthcare and legal services, where confidentiality is paramount. The objective is to develop a secure, biometric-enabled file-sharing system using ECIES and ChaCha20 encryption to ensure user authentication, protect data confidentiality, and evaluate encryption performance in realtime. In traditional manual file-sharing systems, sensitive documents are exchanged via physical media (e.g., USB drives) or unsecured email. Access is typically controlled through basic passwords or by physical custody. There is no integrated encryption mechanism, and files remain vulnerable to unauthorized access, theft, or loss. Authentication relies on easily compromised credentials, and there is no performance benchmarking for applied cryptographic methods. Manual systems suffer from weak authentication, relying solely on passwords that can be guessed or stolen. File transmission is unsecured, and once data is leaked, it’s irretrievable. Access control is poorly enforced, and no encryption ensures content confidentiality. Furthermore, there’s no mechanism to track or benchmark encryption performance, limiting scalability and trust in sensitive environments. The research aims to overcome limitations of manual systems by introducing multi-factor authentication using fingerprint hashing and integrating real-time encryption performance evaluation The proposed system integrates ECIES and ChaCha20 to provide a secure, performance-optimized file-sharing platform. Upon registration, users submit SHA-256–hashed usernames and fingerprint images to enable biometric login. Uploaded files are encrypted using ECIES with the user’s public key to ensure only authorized decryption. Simultaneously, ChaCha20 encrypts the file to benchmark speed, revealing that it outperforms ECIES for large files. All metadata—owner, filename, and access type—is stored in a MySQL database, and access is controlled per file. Real-time graphing with Matplotlib compares encryption speeds, allowing administrators to make informed choices. This hybrid approach ensures strong security and operational efficiency

Keywords : Cloud data security, privacy preservation, encryption, impersonation prevention, robust authentication, biometric security, ECIES, ChaCha20, secure file sharing, anonymous access, cryptographic framework

Author : N. Arpitha, S. Deepika, K. Sujith, Ravula Hari Krishna, Korra Sai Kiran

Title : ENHANCING CLOUD DATA SECURITY AND PREVENTING IMPERSONATION ATTACKS

Volume/Issue : 2025;02(08)

Page No : 613-624

The growing demand for efficient ride-sharing and food delivery services has necessitated innovative solutions to accurately predict and manage driver demand. These services rely heavily on timely driver availability to meet customer expectations and ensure operational efficiency. Historically, driver demand in such industries has been addressed using manual forecasting methods and basic statistical models. However, these approaches often fall short in capturing the dynamic and unpredictable nature of demand—especially during peak hours, special events, or sudden environmental changes. Traditional systems, limited by their inability to incorporate real-time data, have led to delays, customer dissatisfaction, and inefficient resource allocation. Before the advent of AI, these challenges were tackled using rudimentary scheduling systems, fixed staffing policies, and reactive strategies that failed to adapt to rapidly changing conditions. These shortcomings, combined with the rapid growth of on-demand services, highlighted the urgent need for robust, data-driven solutions. This research is motivated by the limitations of traditional forecasting methods and aims to leverage advanced machine learning techniques to improve demand prediction. Inspired by the success of AI-driven systems in addressing complex, non-linear problems across various industries, the proposed system utilizes ensemble learning techniques and time-series analysis to forecast driver demand accurately. By incorporating historical trends, external variables such as weather and events, and real-time data, the system delivers precise forecasts that enable proactive resource management. This research not only enhances operational efficiency but also improves customer satisfaction by reducing wait times and optimizing driver allocation in ride-sharing and food delivery platforms.

Keywords : Driver Demand Prediction, Time Series Analysis, Ensemble Learning, Machine Learning, Operational Efficiency

Author : S. Puneeth, A. Aravind Goud, M. Sreeram, K. Thirumalesh, K. Venkatesh

Title : DRIVER DEMAND PREDICTION USING MACHINE LEARNING AND TIME SERIES MODELS

Volume/Issue : 2025;02(08)

Page No : 625-634