http://localhost:8080/xmlui/handle/123456789/3419 Tue, 23 Jun 2026 06:28:59 GMT 2026-06-23T06:28:59Z http://localhost:8080/xmlui/handle/123456789/3551 Title: A Musical Composition Assistant System using LSTM Authors: Fegade, Poonam G.; Jagtap, Trupti Abstract: Universally, music is one of the elements that create harmony in this world. Traditionally playing music or creating it has always been seen as a manual task. Music can be created with the help of instruments, voices and sounds. In the Modern Era of technology and AI, we can automate this process. For this, we must understand the basic structure of how music forms and view it scientifically. For the proposed system, we are using RNN-LSTM algorithm to generate music from given sample inputs. A Musical Composition Assistant System (MCAS) involves transformation of music scores into time series representation, encoding the music. A model is designed to execute this algorithm where data is represented with the help of musical instrument digital interface (MIDI) file format for easier access and better understanding. Pre-processing of data before feeding it into the model, revealing methods to read, process and prepare MIDI files for input are used. This system creates good music pieces in MIDI format with given input. The proposed system uses Flask API to interact with the frontend Fri, 15 Jul 2022 00:00:00 GMT http://localhost:8080/xmlui/handle/123456789/3551 2022-07-15T00:00:00Z http://localhost:8080/xmlui/handle/123456789/3550 Title: Motions Detection of Senior Citizen Using Machine Learning Authors: Bagal, V. C.; Hanpode, Pratik Abstract: Recognizing human motions is important from security point of view at any level and scenario. As there are plenty of human motions in a fraction of second, so classification of each motion is challenging task in real world. A Human activity Recognition System recognizes the Shapes and or orientation depending on implementation to task the system into per forming some job. Movement is a form of nonverbal information. A person can make numerous movements at a time. The proposed work aims to detect the movement and actions of a person using image detection methodology. Human activity recognition (HAR) aims to recognize activities from a series of observations on the actions of subjects and the environmental conditions. The vision-based HAR research is the basis of many applications including video surveillance, healthcare, and human-computer interaction (HCI). The proposed work is suitable to identify objectionable human motions of senior citizen who live alone at home. Thu, 15 Sep 2022 00:00:00 GMT http://localhost:8080/xmlui/handle/123456789/3550 2022-09-15T00:00:00Z http://localhost:8080/xmlui/handle/123456789/3425 Title: Detection of Anomalies using Local Outlier Factor and Isolation Forest algorithm Authors: Nimani, Sunny; Khairnar, Mahesh; Khele, Prathamesh; Patil, Vishal; Banait, S.S. Abstract: : Data generated from smart devices or applications are in time-series format, in which information is recorded for each specific time. Anomalies in log data refer to certain patterns or points in data that deviate from average data. Anomaly detection is concerned with identifying data patterns that deviate remarkably from the expected behavior. This is an important research problem, due to its broad set of application domains, from data analysis to e-health, cybersecurity, predictive maintenance, financial fault prevention, and industrial automation. Efficiency of Local Outlier Factor Algorithm, Isolation Forest Algorithm is compared. Testing dataset is obtained from Indian Council of Medical Research (ICMR) and credit card company transactional data. Sun, 05 Jun 2022 00:00:00 GMT http://localhost:8080/xmlui/handle/123456789/3425 2022-06-05T00:00:00Z http://localhost:8080/xmlui/handle/123456789/3424 Title: Underwater Marine Life Detection Using Image Processing Authors: Joshi, Yash; Desale, Rutik; Dixit, Sairaj; Jadhav, Malhar; Mahajan, Monali Abstract: Marine life research and computer technology have been utilized in tandem for the study of aquatic ecosystems and the analysis of ocean floors throughout the last few decades. Few modern solutions have been offered in this field in recent years. The work in object detection and recognition based on machine learning models have given good information about the surroundings and behavior of marine ecosystems. These models are complex in usage, they often rely on the information source from multiple data forms. The major task is to remove the high impurities in underwater images as the noise removal process is difficult. The image extraction is carried out using darknet which helps in proper object detection. Due to this, the actual applications and study of marine life is realized easily. A suitable environment will be created so that machine learning algorithms such as YOLO will be used to detect and recognize the animals under the ocean with the help of image processing Mon, 20 Jun 2022 00:00:00 GMT http://localhost:8080/xmlui/handle/123456789/3424 2022-06-20T00:00:00Z