International Journal on Recent and Innovation Trends in Computing and Communication
Not a member yet
8613 research outputs found
Sort by
Cosine Modified K-Means and Neural Network for Classification of Images
Get PDFA significant amount of data transfer happens daily through the internet in the form of images, PDFs, and videos. This exchange rate has increased tremendously during the COVID-19 pandemic. However, data transfer consumes a lot of internet bandwidth. It can be reduced significantly if there was a way to determine whether an image is compressed or non-compressed. While much research has been done on image compression in modern photography, detecting whether an input image is compressed or uncompressed has not been studied. This research aims to develop a new algorithm to classify images as compressed or uncompressed. The first step is to propose a new clustering method that uses two random centroids based on randomly selected pixels from the image. The Euclidean distance of each pixel from the randomly selected centroids is used to perform clustering. If clustering fails, then the cosine similarity method is used to perform clustering. This method is called the Cosine modified K-Means method. The SURF feature detection method is used to find the features of each image. Based on these extracted features, a neural network is trained. To test the algorithm, a random image is selected and passed through the network. The algorithm can classify the image as compressed or uncompressed. Precision, recall, classification accuracy, and error rate are calculated to evaluate the performance of the method
Enhanced Hub Location Routing with Branch-and-Cut Methods and Simplified Mathematical Models
Get PDFThis article presents an advanced approach to the hub location routing problem, focusing on the optimal placement of hub nodes and the allocation of spoke nodes. We introduce a novel branch-and-cut method combined with a new simplified mathematical model incorporating valid inequalities. This hybrid approach aims to enhance solution quality and computational efficiency. Our proposed method integrates a learning mechanism to guide local searches, leveraging dual information from Lagrangian relaxation. Computational experiments validate the effectiveness of the proposed method
Machine Learning Algorithms in Cloud Manufacturing - A Review
Get PDFCloud computing has advanced significantly in terms of storage, QoS, online service availability, and integration with conventional business models and procedures. The traditional manufacturing firm becomes Cloud Manufacturing when Cloud Services are integrated into the present production process. The capabilities of Cloud Manufacturing are enhanced by Machine Learning. A lot of machine learning algorithms provide the user with the desired outcomes. The main objectives are to learn more about the architecture and analysis of Cloud Manufacturing frameworks and the role that machine learning algorithms play in cloud computing in general and Cloud Manufacturing specifically. Machine learning techniques like SVM, Genetic Algorithm, Ant Colony Optimisation techniques, and variants are employed in a cloud environment
Rare And Popular Event-Based Co-Located Pattern Recognition in Surveillance Videos Using Max-Min PPI-DBSCAN And GREVNN
Get PDFCo-located pattern recognition is the process of identifying the sequence of patterns occurring in surveillance videos. In greater part of the existing works, the detection of rare and popular events for effective co-located pattern recognition is not concentrated. Therefore, this paper presents the automatic discovery of the co-located patterns based on rare and popular events in the video. First, the video is converted to frames, and the keyframes are preprocessed. Then, the foreground and background of the frames are estimated, and the rare and popular events are grouped using Maximum-Minimum Pixel-Per-Inch Density-Based Spatial Clustering of Applications with Noise (Max-MinPPI-DBSCAN). From the grouped image, the object detection and mapping are done, and the patch is extracted from it. Next, the edges are detected and from that, for the moving objects, motion is estimated by the Kullback-Leibler Kalman Filter (KLKF). Also, for non-moving objects, the objects/persons are tracked. From the motion estimated and tracked data, time series features are extracted. Then, the optimal features are selected using the Dung Beetle State Transition Probability Optimizer (DBSTPO). Finally, the co-located pattern is classified using a Generalized Recurrent Extreme Value Neural Network (GREVNN), and the alert message is given to the authorities. Hence, the proposed model selected the features in 53239.44ms and classified the event with 99.0723% accuracy and showed better performance than existing works
From Malaria to Dengue: A Comprehensive Review of Disease Transition in Rajasthan
Get PDFRajasthan, a historically malaria-prone state in India, is undergoing a profound epidemiological shift, with dengue emerging as a significant public health challenge while malaria incidence demonstrates a commendable decline. This comprehensive review synthesizes existing literature, incorporating illustrative data, to explore the dynamics, drivers, and implications of this disease transition. We examine the historical epidemiology of malaria, noting its peak burdens in the mid-20th century and its subsequent decline, exemplified by an over 95% reduction in annual parasite incidence (API) from 2000 to 2020. Concurrently, we trace the dramatic rise of dengue, with reported cases in Rajasthan escalating from a few hundred annually in the early 2000s to over 15,000 cases in epidemic years like 2017 and 2021, predominantly concentrated in urban centers like Jaipur. The review details the complex interplay of environmental, climatic (e.g., changing rainfall patterns and rising minimum temperatures), socio-economic (e.g., rapid urbanization at an average annual rate of over 3%), and public health factors contributing to this shift. It also assesses current strategies, identifies gaps in surveillance and control, and discusses challenges posed by co-circulation and diagnostic complexities. Ultimately, this paper underscores the urgent need for integrated, adaptive, and proactive vector-borne disease control programs in Rajasthan that account for the evolving epidemiological landscape and prepare for future challenges posed by climate change and demographic shifts
A Comprehensive Review on Machine Learning Based Models for Healthcare Applications
Get PDFAt present, there has been significant progress concerning AI and machine learning, specifically in medical sector. Artificial intelligence refers to computing programmes that replicate and simulate human intelligence, such as an individual's problem-solving capabilities or their capacity for learning. Moreover, machine learning can be considered as a subfield within the broader domain of artificial intelligence. The process automatically identifies and analyses patterns within unprocessed data. The objective of this work is to facilitate researchers in acquiring an extensive knowledge of machine learning and its utilisation within the healthcare domain. This research commences by providing a categorization of machine learning-based methodologies concerning healthcare. In accordance with the taxonomy, we have put forth, machine learning approaches in the healthcare domain are classified according to various factors. These factors include the methods employed for the process of preparing data for analysis, which includes activities such as data cleansing and data compression techniques. Additionally, the strategies for learning are utilised, such as reinforcement learning, semi-supervised learning, supervised learning, and unsupervised learning. are considered. Also, the evaluation approaches employed encompass simulation-based evaluation as well as evaluation of actual use in everyday situations. Lastly, the applications of these ML-based methods in medicine pertain towards diagnosis and treatment. Based on the classification we have put forward; we proceed to examine a selection of research that have been presented in the framework of machine learning applications within the healthcare domain. This review paper serves as a valuable resource for researchers seeking to gain familiarity with the latest research on ML applications concerning medicine. It aids towards the recognition for obstacles and limitations associated with ML in this domain, while also facilitating the identification of potential future research directions
Personality Prediction based on Myers Briggs type Indicator Using Machine Learning
Get PDFIn this study, we leverage a combination of machine learning algorithms, including classification and regression models, along with natural language processing techniques, such as NLP and spacy, to predict user personality types from their social media posts. We focus on utilizing the Myers-Briggs Type Indicator (MBTI) to identify a user's unique personality among sixteen possible types [1]. This research aims to establish a correlation between individuals' social media content and their personality traits. Our approach involves extensive preprocessing of textual data, employing techniques like text tokenization, regular expressions, lemmatization, sentiment analysis, and part-of-speech tagging, followed by dimensionality reduction [2]. We evaluate several machine learning models, including logistic regression, SVM, Naive Bayes, lasso regression, and random forest classifiers, with logistic regression delivering the most accurate results. We deploy this trained model on a web page connected to a Flask app, allowing users to input a brief description of themselves and receive their predicted personality type. This research explores the intersection of text analysis and personality prediction, shedding light on the hidden dimensions of human personality revealed through digital traces in the age of social media [4]
Solar and Dynamo Powered Ev Using Fingerprint Authentication
Get PDFNonrenewable energy sources are already interfering with the availability of electricity. Today, the primary emphasis is on the generation of electricity from renewable sources. This proposal proposes a feasible design solution in the shape of a user-friendly three-wheeler. A fingerprint-accessible electric vehicle. This electric vehicle uses solar power and a self-changing dynamo to charge its battery. The battery serves as the vehicle's power source.
A microcontroller serves as the system's primary controller. It is connected to a fingerprint module, a relay, the car ignition, and an LCD. The Microcontroller reads the input from the fingerprint module and, if it is legitimate, grants access to the ignition system. The status will be displayed on the LCD.
In addition, we're adding three more sensors: speed, temperature, and voltage. Lithium-ion batteries can be harmful if not used within the safety-operated area (SOA). To avoid this, we calculate the temperature of the lithium-ion battery when it exceeds 49.9?. It will make a sound through the buzzer. This can prevent damage to the battery and the person driving the vehicle. Voltage Sensor is designed to monitor battery charging and discharging time, which is displayed on the LCD. A speed sensor is used to determine the speed of the vehicle
Load Balancing Factor of DAG Based BNP Scheduling Algorithms
Get PDFEffective scheduling of applications is crucial for achieving optimal performance in uniform computing environments. The scheduling problem is known to be NP-complete in both general and specific cases. Given its paramount importance, various BNP algorithms, including HLFET, MCP, ETF, and DLS, have been extensively explored, primarily designed for parallel processing systems. This study evaluates the performance of these four algorithms utilizing a Direct/Arbitrary Task Graph (DAG) comprising 11 tasks, focusing on key performance parameters such as efficiency and load balancing. The MCP algorithm demonstrates superior efficiency, while HLFET excels in terms of load balancing
Assessing the Performance of Handcrafted Features for Human action Recognition
Get PDFRecognition of Human action such as running, punching, bending, kicking etc. plays an vital role in futuristic applications like intelligent video surveillance, health care monitoring, robotics, smart automation system, computer gaming etc. This field relies on various approaches based on hand crafted features like PCA, HOG, LBPH, DWT, STIP, SWF, SWFHOG and deep learning techniques like CNN, RNN and their variants. Though many approaches have been proposed and implemented by researchers, the literature survey suggests that a detailed understanding of the approaches and a comparison of advantages and limitations is required to develop more accurate action recognition method. This paper focuses on this issue and gives detailed analysis of results obtained by implementing algorithms on standardize open source datasets of varying complexity namely Weizmann, KTH, UT Interaction and UCF sports. The results are compared based on the classification accuracy as it is one of the performance measure for checking reliability of the method. The comparison shows that, SHFHOG feature gives the best classification accuracy as compared to other handcrafted features and also outperforms the simple CNN