Volume 12 | Issue 5 |

 Abstract

Sign Language Recognition and Translation to English and Marathi is a project that aims at enhancing communication between the deaf and the hard of hearing (DHH) community with the hearing populace. Sign language, which is the primary means of communication among DHH, is often a barrier to various aspects such as education, health care, employment and social interactions. This project seeks to eliminate this communication barrier using state-of-the-art technology by recognizing sign-language gestures for spoken and written language in English and Marathi--two majorly spoken languages. The project serves an essential purpose of improving communication between the Deaf-Hard of Hearing (DHH) community and the general population. A primary form of communication used by DHH is sign language, which remains problematic in different areas such as education, health care, jobs or even social interaction. In order to solve this problem, cutting-edge technologies are utilized in order to create an innovative solution which can detect sign language gestures and translate them into spoken or written forms in both English as well as one of the major Indian languages - Marathi.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Machine Learning, MediaPipe, OpenCV, LSTM Neural Network, Sign Language

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

One of the most vital senses for any individual is the ability to see. Unfortunately, millions of people worldwide grapple with vision impairments, which pose significant challenges in terms of communication and accessing information. This struggle often hinders their ability to navigate safely and independently. To address this issue, the proposed work seeks to transform the visible world into an auditory one. This transformation will be achieved by harnessing real-time object detection technology, empowering individuals with vision impairments to move autonomously without external assistance. Through the application of image processing and machine learning, the program can swiftly identify objects in real time using the camera and convey their locations to blind users through voice output. The inability to differentiate between objects has given rise to numerous problems, and this innovative technology aims to provide a solution.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Object Detection, Android Application, YOLO, CNN (Convolutional Neural Network), Visually Impaired people, Computer Vision, Algorithms

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

An online platform is under development for a college's placement management system, aimed at optimizing the recruitment process and fostering better communication among students, educational institutions, and potential employers. This system will serve as a centralized hub for managing student information, encompassing personal details, academic records, technical competencies, and career aspirations. Additionally, it will enable students to register online for placement opportunities, apply for relevant positions, and monitor their application progress seamlessly. Employers will gain access to a dedicated portal to search for suitable candidates, schedule interviews, and engage with students and placement officers efficiently. Furthermore, the system's implementation will advance toward a paperless environment by digitizing the entire placement procedure, thereby reducing paperwork and promoting environmental sustainability.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Web development, Authorization, Student, Admin, TPO, College

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Fashion holds significant sway in our everyday lives, serving as a mirror of our personal style and identity. Yet, navigating the world of fashion and personal style presents challenges, particularly in choosing outfits that flatter individual body types and shapes. This process often proves daunting and time-consuming, leading to indecision and a lack of confidence in one's appearance. To tackle these hurdles, this study aims to develop an ML-based Outfit Suggestion System. Introducing an innovative approach, this system harnesses machine learning methodologies, including deep learning, computer vision, and natural language processing. By scrutinizing an extensive dataset encompassing clothing items and body shape attributes, the system furnishes tailored outfit recommendations designed to suit individual body types and shapes. This research marks a notable stride in the evolution of fashion recommendation systems, offering a promising avenue for fashion enthusiasts seeking personalized outfit guidance across varied contexts.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Fashion Recommendation, Clothing Recommendation, Machine Learning, Fashion Dataset, Body Shape Analysis, Body Type Analysis, Body Types, Fashion

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Choosing outfits that complement our body types can be challenging and time-consuming, leading to uncertainty and a lack of confidence. This study aims to address these challenges by developing a Machine Learning- based Outfit Suggestion System. This innovative system utilizes various machine learning techniques, such as deep learning, computer vision, and natural language processing, to analyze a vast dataset containing clothing items and body shape attributes. By doing so, it provides personalized outfit recommendations tailored to individual body types. This research represents a significant advancement in fashion recommendation systems, offering a promising solution for fashion enthusiasts seeking personalized outfit suggestions across different contexts.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Style Advice, Apparel Suggestions, AI-driven Fashion Guidance, Apparel Dataset, Physique Assessment, Physique Analysis, Body Structures, Apparel Trends.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Abstract--Cryptocurrency markets exhibit high volatility, making accurate price prediction a challenging task. This paper shows us a good approach to cryptocurrency price prediction using deep learning techniques, specifically Long Short-Term Memory (LSTM) neural networks. The study utilizes historical cryptocurrency data (BTC-USD1.csv) and applies preprocessing techniques to prepare the dataset for model training. The LSTM model is trained on this data to forecast short-term price movements. Findings show how well the model works to forecast cryptocurrency values, giving traders and investors valuable information.The paper concludes with discussions on the implications of the findings and suggestions for future research directions in the field of financial forecasting using deep learning.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Cryptocurrency, Price Prediction, Deep Learning, LSTM, Neural Networks, Financial Forecasting

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In today's dynamic urban environments, efficient cab management systems are crucial for seamless transportation and passenger satisfaction. This paper presents the design and implementation of a cab management system built with Flutter for a mobile frontend and Firebase for a robust backend. Our primary focus lies on optimizing the driver-rider matching process to ensure timely cab allocation and minimize wait times. The paper details the system architecture, outlining how Flutter's cross-platform capabilities create a user-friendly mobile application for both riders and drivers. Firebase's real-time functionality is leveraged to facilitate efficient communication and data exchange between app users and the backend system. We delve into the core aspects of our driver-rider matching algorithm, explaining how it considers various factors to optimize connections. This may include factors like driver location, rider destination, and real- time traffic conditions (if implemented). By implementing this cab management system, we aim to demonstrate the effectiveness of Flutter and Firebase in building a scalable and optimized solution for matching cab drivers with riders. The paper concludes by discussing the potential benefits of this system for transportation service providers and riders alike, emphasizing improved efficiency and user satisfaction.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Cab management, Flutter, firebase,Driver-rider matching, Optimization, Real-time communication,alability, Efficiency, User satisfaction, Transportation

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In the vast landscape of machine learning, the emergence of adversarial threats has cast a shadow over the reliability and security of deployed models. With the proliferation of sophisticated attacks aimed at undermining the integrity of machine learning systems, the imperative for robust defenses has never been more pronounced. Against this backdrop, this paper embarks on a comprehensive journey through the adversarial landscape, surveying the myriad threats and safeguards that define the contemporary discourse in machine learning security. Under the banner of "Navigating the Adversarial Landscape," this survey endeavors to shed light on the intricate interplay between adversarial attacks and defensive strategies. By analyzing the life structures of ill- disposed dangers and examining the viability of existing protections, this try looks to outfit per users with a nuanced comprehension of the difficulties and open doors intrinsic in defending AI frameworks. As we embark on this expedition, we delve into the nuanced nuances of adversarial attacks, encompassing a spectrum of techniques ranging from subtle perturbations to outright manipulations. From white-box to black-box attacks, and from transfer to physical assaults, we unravel the diverse tactics employed by adversaries to subvert machine learning systems. However, amidst the looming specter of adversarial threats, glimmers of hope emerge through the pursuit of robust defense mechanisms. Through adversarial training, robust optimization, and certified defenses, among other strategies, researchers endeavor to fortify machine learning models against adversarial incursions. Ultimately, the quest to navigate the adversarial landscape represents not only a technical challenge but also a moral imperative in safeguarding the integrity and trustworthiness of machine learning systems.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Machine Learning Security, Robustness, Vulnerabilities, White-Box Attacks, Black-Box Attacks, Transfer Attacks, Physical Attacks, Defense Mechanisms, Adversarial Training, Robust Optimization, Feature Denoising, Certified Defense

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In the ever-evolving landscape of music consumption, the development of intelligent recommendation systems has become imperative to enhance user experience. This research paper introduces a pioneering approach to music recommendation by integrating advanced Convolutional Neural Networks (CNN) with face expression recognition. The proposed system aims to personalize music suggestions by analyzing users' facial expressions, extracting emotional cues, and aligning them with the corresponding auditory preferences. The convolutional neural network component of the system is designed to learn intricate patterns and features from music spectrograms, capturing both the audio content and underlying emotional nuances. Simultaneously, facial expression recognition technology is employed to discern users' emotional states during music listening sessions. By fusing these two modalities, our system strives to create a holistic understanding of users' preferences, considering both explicit musical features and implicit emotional responses. To achieve this integration, we leverage machine learning architectures for music analysis and facial expression recognition. A wide variety of facial expressions and musical genres are included in the dataset that the model is trained on. Additionally, the research explores the challenges and opportunities associated with combining these distinct modalities, such as data preprocessing, feature extraction, and model fusion. This research contributes to the ongoing discourse on the fusion of multimodal technologies for more nuanced and personalized recommendation systems, paving the way for innovative applications in the intersection of music and affective computing.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Music, CNN, Expression, Feature Extraction

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

We propose a new approach for playing music automatically using facial emotion Current methods often involve manually recording music, using computer-based tools, or classifying sounds. Instead, we recommend manually changing the way you rank and play. We use convolutional neural networks for emotion recognition. Pygame and Tkinter are available for down load. Our proposed method will reduce the calculation time as well as reduce the total cost of obtaining results and building the system, thus improving the overall accuracy of the system. The testing was done on the FER2013 dataset. Capture face with built in camera. Feature extraction is performed on facial images to direct emotions such as happiness, anger, sadness, surprise and neutrality

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Face Recognition, Feature extraction, Emotion detection, Convolutional Neural Network, Pygame

  License

Creative Commons Attribution 4.0 and The Open Definition