volume-2 CSIT issue 2
RUNGTA INTERNATIONAL OF JOURNAL OF COMPUTER SCIENCE AND INFORMATION TECHNOLOGY
RUNGTA INTERNATIONAL OF JOURNAL OF COMPUTER SCIENCE AND INFORMATION TECHNOLOGY
• Volume 2 (Jan 2025- Dec 2025)
Issue 2
1. Exploring Technological Revolution and Adoption for Blockchain Technology with its Potential Fit for Sustainable Societal and Environmental Changes
Aradhana Sahu1,#, Bhawna Janghel2, Shaziya Islam3, Shweta Bandhekar4 , Tripti Sharma5, Huma Khan6Author Affiliations
123456 Department of Computer Science and Engineering, Rungta College of Engineering and Technology, Bhilai CG 490024
ABSTRACT:
Blockchain technology has become a revolutionary advancement with broad ramifications for many industries. It is an open-source, decentralized digital ledger that securely records and verifies network transactions. This paper reviewed and examines the foundational features of blockchain technology and how they have a significant influence on technological adoption. The blockchain has significant effects on societal and environmental changes that are sustainable. It addresses important issues such as patient privacy, data security, and interoperability. Blockchain enables secure medical record exchange and management, maintaining data integrity and giving healthcare professionals access to patient data in real time. Additionally, it makes clinical trials, drug tracing, and the detection of fake pharmaceuticals easier. First and foremost, blockchain has changed the financial sector. Blockchain-based cryptocurrencies have replaced conventional fiat money as a viable alternative. It enables effective international payments, unbanked people’s financial inclusion, and the potential for decentralized finance (DeFi) applications. Additionally, Innovation in many different areas, such as energy, supply chains, real estate, intellectual property, and more, has been spurred by blockchain technology. Because of its decentralized structure, cryptographic security, and trustless environment, entrepreneurs and developers have been motivated to consider new ideas and develop original applications. In conclusion, blockchain technology has a huge and diverse impact on how technologies are adopted. It challenges conventional systems, increases productivity across industries, and enhances security.
Keywords: Keywords: Interoperability; Decentralization; Security; Digital Assets; Secure Sharing; Healthcare; Cryptocurrency
2. AI-Powered Autonomous Robotic System for Intelligent Vehicle Insurance Assessment and Fraud Detection
Onika Parmar1,#, Shweta Pandey1, Rani Namdev Sinha1, Roushan Kumar1 , Madhuri Gupta1, Padmavati Shrivastava1Author Affiliations
1 Department of Computer Science Engineering (Artificial Intelligence and Machine Learning),Rungta College of Technology, Bhilai, India-490024
ABSTRACT:
This paper presents an advanced, field-deployable AI-driven robotic system engineered to revolutionize vehicle loss assessment and fraud detection. The system integrates an articulated robotic arm with high-resolution visual, LiDAR, infrared, and thermal sensors to capture multidimensional damage data, while IoT-enabled modules retrieve real-time telemetry from vehicle systems. Utilizing encrypted data processing and machine learning algorithms, the system performs causal and spatial analysis to identify inconsistencies between claimed and observed accident data. A block chain-based ledger ensures the immutability and transparency of all transactions, enhancing trust and regulatory compliance. Through autonomous navigation, real- time decision-making, and predictive analytics, the system not only streamlines the insurance workflow but also elevates its precision and reliability. This intelligent robotic platform represents a significant advancement toward digitizing and automating insurance operations, reducing fraud, and ensuring equitable outcomes for all stakeholders.
Keywords: Machine learning, thermal sensing, blockchain, multi-sensor fusion, real- time data encryption, predictive analytics
3. Emotion-Aware Social Media Feed with Real- Time Emotion Recognition to Enhance Mental Health
Pinki Patra1,#, Richa Sahu1, Sonal Yadav1, D Mohini1, Chanchala Patel1 , Mahendra Kumar Soni1Author Affiliations
1 Department of AI& AIML,Rungta College of Engineering & Technology, Bhilai, India-490024
ABSTRACT:
Social media now plays a crucial role in everyday life by offering channels for exchanging information, communicating, and expressing oneself. However, users’ mental health can be severely impacted by excessive and uncontrolled exposure to emotionally upsetting or negative content, which can result in social isolation, stress, anxiety, and despair. Conventional social media platforms frequently lack emotional intelligence and are ill-prepared to react to users’ emotional states in a dynamic manner. In order to provide a customized and emotionally responsive user experience, this project suggests an Emotion-Aware Social Media Feed that makes use of real-time emotion recognition. The system can determine users’ emotions while they engage with material by combining technologies including speech tone detection, natural language processing (NLP), and facial expression analysis. The platform prioritizes and curates’ information based on these emotional insights, limiting exposure to potentially dangerous content and encouraging postings that are uplifting, relaxing, or motivational when distressing indications are identified. Promoting emotional awareness, improving digital well-being, and offering early help or treatments when needed are the objectives. Furthermore, the system prioritizes ethical aspects such as fairness across various demographic groups, secure data handling, and informed user permission. In order to create better social media connections, this project intends to turn passive content consumption into an emotionally intelligent and mentally supporting experience by fusing artificial intelligence with affective computing concepts
Keywords: Emotion recognition, mental health, affective computing, personalized social media, digital well-being.
4. Transformer Models for Time-Series Forecasting: A Review of Recent Advances and Applications in Energy and Finance
Shubh Raj Gupta1, #, Subham Pandit2, Shivam Sahu3, Aradhana Sahu4 , Bhawna Janghel5Author Affiliations
145 Department of Computer Science and Engineering, Rungta College of Engineering and Technology, Bhilai CG 490024
2 Department of Computer Science & Engineering (AI ML), Rungta College of Engineering and Technology, Bhilai CG 490024
3 Department of Computer Science and Engineering (Cyber Security), Rungta College of Engineering and Technology, Bhilai CG 490024
ABSTRACT:
Precise time-series prediction is essential in applications ranging from finance to energy, where the decision heavily depends on the capacity to forecast changing temporal patterns. Popular models such as ARIMA, Prophet, and LSTM have performed satisfactorily but are usually not adequate in capturing long-range dependencies and coping with irregular or multivariate series. Transformer- based architecturs, initially used for natural language processing, have recently been found useful for modeling long-term dependencies and scaling well to large datasets. This has provided new possibilities for their use in time-series forecasting tasks. This review explores recent developments and new trends in the application of Transformer models to time series forecasting, with specific focus on their use in energy systems and financial markets. We review a number of specialized variants of the Transformer—Informer, Autoformer, and Temporal Fusion Transformer (TFT)—and how they specifically handle domain-specific forecasting difficulties like seasonality, multi-horizon forecasting, and high-frequency noise. In addition, we offer a comparative analysis of primary model architectures, learning processes, benchmarking datasets, and performance metrics widely utilized in these applications. We also discuss existing limitations, such as model interpretability, sparsity of data, and real-time deployment restrictions, and highlight research opportunities in the future, notably in enhancing efficiency and domain adaptation. Lastly, we briefly sketch the potential of these technologies to serve towards Sustainable Development Goals, in particular SDG 7 (Affordable and Clean Energy) and SDG 8 (Decent Work and Economic Growth).
Keywords: Breast cancer; ultrasound; explainable artificial intelligence; deep learning
5. Sentiment Analysis of Amazon Product Reviews Using VADER: Natural Language Processing Techniques
Nikita Sinha1, #, Shaziya Islam2Author Affiliations
1 MTech Computer Science & Engineering, Rungta College of Engineering and Technology, Bhilai, Chhattisgarh, India-490024
2 Associate professor, Department of Computer Science & Engineering, Rungta College of Engineering and Technology, Bhilai, Chhattisgarh, India-490024
ABSTRACT:
In this age of digital transformation, consumer reviews have evolved into the most important source for gauging public opinion, which can ultimately affect consumer purchasing behavior and production. This study aims to utilize sentiment analysis in deriving useful information from Amazon product reviews based on the VADER (Valence Aware Dictionary and Sentiment Reasoner) sentiment analysis tool. VADER is a rule- and lexicon-based model particularly developed to assess sentiments in social media and product review text. In the current research, we process a huge dataset of text reviews systematically, calculate sentiment polarity scores, and label them as positive, negative, or neutral. Keyword filtering based on keywords is also employed to investigate consumer attitude on certain product features like battery life, quality, and value. The results indicate that there is a high correlation between sentiment scores and user ratings, thus serving as evidence that VADER can capture consumers’ perception adequately. In this study we also utilize state-of-the-art visualization methods—e.g., violin plots, histograms, and pie charts— to clarify sentiment distributions and rating tendencies. Overall, the results illustrate the effectiveness of VADER, a quick and lightweight tool for opinion mining in e-commerce, and valuable insights for manufacturers, retailers, and policymakers trying to maximize user experience and product quality.
Keywords: Sentimental Analysis; Natural Language Processing; Amazon Review; Sentiment Visualization; Keyword-Based Sentiment Filtering; VADER