Credit card fraud is a serious criminal offense. It costs individuals and financial institutions billions of dollars annually. According to the reports of the Federal Trade Commission (FTC), a consumer protection agency, the number of theft reports doubled in the last two years. It makes the detection and prevention of fraudulent activities critically important to financial institutions. Machine learning algorithms provide a proactive mechanism to prevent credit card fraud with acceptable accuracy. In this paper Machine Learning algorithms such as Logistic Regression, Naïve Bayes, Random Forest, K- Nearest Neighbor, Gradient Boosting, Support Vector Machine, and Neural Network algorithms are implemented for detection of fraudulent transactions. A comparative analysis of these algorithms is performed to identify an optimal solution.

• Credit Card Fraud Detection
• Machine Learning Algorithms

In today’s digital age, vast information exists in non-digital forms like books and handwritten notes. Optical Character Recognition (OCR) digitizes such content from images using image processing and machine learning. Our system preprocesses images, removes noise, and applies OCR algorithms. The extracted text is saved, offering efficient digitization and enhancing various applications.

• Digitization Solutions
• Image Processing Algorithms
• OCR Technology

Underwater images often suffer from light scattering, absorption, and reflection, causing reduced visibility and color distortion. We propose a two-stage method for underwater image dehazing. First, we estimate the transmission map using the dark channel prior, then apply color correction for improved color balance. Our approach outperforms others in visual quality and quantitative metrics.

• Color Correction
• Dark Channel Prior
• Underwater Image Dehazing

Mobile phones, used by all ages, enhance convenience. Smartphones offer diverse functions Choosing the right one is tough due to myriad options. Users rely on reviews and prices for decisions.  Python web app aims to classify and rank smartphone features based on user preferences. It employs machine learning algorithms to analyze smartphone data and refine recommendations with user feedback. The user-friendly interface allows input of preferences for camera quality, battery life, display resolution, etc. By processing a dataset of smartphone features and user ratings,  apply regression analysis, clustering, and decision trees. It offers personalized smartphone recommendations, empowering users to make informed choices and enhance their smartphone experience.

• Machine Learning Recommendations
• Smartphone Selection
• User Preferences

Scholar placement is a crucial aspect of academic institutions, impacting admissions. Institutions enhance placement departments to bolster this process. This paper analyzes historical student data to predict placements, presenting a recommendation system employing Naive Bayes and K Neighbors algorithms to increase efficiency. The system aids in identifying potential students for skill improvement.

• Academic Placement
• Predictive Analytics
• Recommendation System

In recent years, rising interest in online education, including MOOCs and SPOCs, has brought forth challenges like student engagement and performance prediction. This review covers cutting-edge research on using machine and deep learning to predict online learners’ outcomes, categorizing features, strategies, and evaluation metrics while addressing challenges and limitations.

• Learner Performance Prediction
• Machine Learning in Education
• Online Education

Generating realistic faces from sketch images or textual descriptions is a fundamental challenge in computer vision due to the limited facial details in sketches. this has a face hallucination problem. It involves an image translation network utilizing adversarial networks to enhance facial attribute accuracy. It combines sketch images and attribute features perceptually, distinguishing it from most attribute-embedded networks. network comprises a feature extractor and down-sampling/up-sampling networks, using skip-connections to reduce layer complexity while maintaining performance. The discriminator assesses attribute presence in generated faces. This approach outperforms current image translation methods.

• Adversarial Networks
• Computer Vision
• Image Translation

In this study, we introduce a novel approach, the Multi-Stream Adaptive Graph Convolutional Neural Network (MS-AAGCN), for skeleton-based action recognition. It addresses previous issues by dynamically learning graph topology and incorporating second-order skeleton data information. The model’s adaptability enhances generality and accuracy, outperforming existing methods on NTU-RGBD and Kinetics-Skeleton datasets.

• Graph Convolutional Neural Network
• NTU-RGBD Dataset
• Skeleton-based Action Recognition

The Rail Time app offers Indian Railways services and information, facilitating journey planning from any station by any train, preempting service disruptions. It covers station details, train schedules, PNR status, fare, accommodation, and reservations via a user account. Beneficial for tourist bureaus aiding foreign tourists and NRIs.

Modules Description

• Indian RailwaysApp
• Train Schedules
• Travel Planning

The surge in social networking’s popularity has led to a rise in social network mental disorders (SNMDs). These include Cyber-Relationship Addiction, Information Overload, and Net Compulsion. Early detection is crucial. We propose a machine learning approach, SNMDD, leveraging social network data to identify SNMD cases. Our innovative STM model enhances accuracy and scalability. Evaluation involving 3,126 users confirms SNMDD’s promise in identifying potential SNMDs.

• Machine Learning in Healthcare
• Mental Health Detection
• Social Network Disorders