International Journal on Recent and Innovation Trends in Computing and Communication
Not a member yet
8613 research outputs found
Sort by
State of Health Estimation of VRLA Batteries Using Vibration Tools.
Get PDFVRLA batteries are widely accepted primary source of energy for electrical vehicles due to their cost efficiency and will remain in the near future. Despite this advantage, VRLA batteries are prone to acute as well as chronic defects due to continual usage, and they may develop various faults during their use such as reduction of electrolyte, sulphonation, etc. These faults degrade the battery performance. To sustain battery performance we must continuously measure and monitor state of health and state of charge of the battery. Various methods to measure state of charge have been discussed. And to monitor the health of batteries a vibration sensor is attached to the external surface of the cell. Real time data from the vibration sensor is used to non-destructively evaluate the internal condition of vital interfaces inside the cell. Hence pre-emptive measures are taken against any incipient faults and defects to prevent any further damage to the battery. A regular data of the vibration for batteries used in vehicles may prove to be helpful in realizing the need of repair or replacement. To collect this data a set of vibration sensors are used along with a set-up placed around a battery and a microcontroller for data acquisition from the sensors. This data is then plotted on a graph which is then matched with previously made graphs to understand the type defect that has taken place in the battery or the period of its life it is going through
Revisiting Malthusian Trap Theory in Bangladesh: A Time Series Approach to Recognize the Existence of the Trap
Get PDFThe key feature of the Malthusian model is that output per capita is strongly linked to population growth (Olsson, 2012). This paper uses data from 1996 to 2020 to justify the existence of the Malthusian trap in Bangladesh. Data collection includes per capita GNI growth rate (PCGNI GR), industrial growth rate (IND GR), GNI growth rate (GNI GR), and population growth rate (PGR). Here the regression model and cointegration test have been used to justify the relevance of the theory in case of Bangladesh. According to the regression results and the value of normalized cointegrating coefficients it is found that the change in the growth rate of per capita GNI is same directional at a same rate to the change in the growth rate of GNI of Bangladesh. That means, if the growth rate of GNI increases or decreases by 1% then the growth rate of per capita GNI will also increase or decrease to the same percentage. On the other hand, the population growth has a cent percent negative effect on the growth of per capita GNI. This means that, an increase in per capita income due to an increase in total income will be offset by population growth. These results prove that the Malthusian trap theory is quite relevant for Bangladesh. The results also show that the industrial growth has positive influence on the per capita income of Bangladesh, which can break the trap. Therefore, steps must be taken to control population growth and improve industrial sector to break down the trap
Descriptive statistics of Neural Network and Regression Based Results for Short Term Electricity Demand Prediction
Get PDFIn the realm of data analysis and predictive modeling, both neural networks and regression techniques play pivotal roles. This study aims to provide a comparative analysis of the descriptive statistics derived from neural network and regression-based results. Utilizing a dataset representative of real-world scenarios, we explore how these two approaches perform in terms of descriptive measures such as mean squared error, coefficient of determination (R-squared), standard error, and others.
The research involves implementing both neural network and regression models on the dataset and evaluating their performance using various statistical metrics. Through a systematic examination of the descriptive statistics derived from these models, we aim to elucidate the strengths and weaknesses of each approach in capturing the underlying patterns and making accurate predictions. Additionally, we delve into the interpretability aspect, assessing the ease of understanding the results provided by neural networks compared to regression models.
Furthermore, the study investigates the impact of factors such as dataset size, complexity, and feature selection on the performance and descriptive statistics of neural networks and regression techniques. By conducting experiments across different scenarios and datasets, we aim to provide insights into the conditions under which each method excels and where potential limitations lie. The findings of this research contribute to a deeper understanding of the characteristics and capabilities of neural network and regression models in data analysis and prediction tasks. This comparative analysis serves as a valuable resource for researchers, practitioners, and stakeholders seeking to leverage these methodologies effectively in various domains, ranging from finance and economics to healthcare and beyond
BUGOPTIMIZE: Bugs dataset Optimization with Majority Vote Cluster-Based Fine-Tuned Feature Selection for Scalable Handling
Get PDFSoftware bugs are prevalent in the software development lifecycle, posing challenges to developers in ensuring product quality and reliability. Accurate prediction of bug counts can significantly aid in resource allocation and prioritization of bug-fixing efforts. However, the vast number of attributes in bug datasets often requires effective feature selection techniques to enhance prediction accuracy and scalability. Existing feature selection methods, though diverse, suffer from limitations such as suboptimal feature subsets and lack of scalability. This paper proposes BUGOPTIMIZE, a novel algorithm tailored to address these challenges. BUGOPTIMIZE innovatively integrates majority voting cluster-based fine-tuned feature selection to optimize bug datasets for scalable handling and accurate prediction. The algorithm initiates by clustering the dataset using K-means, EM, and Hierarchical clustering algorithms and performs majority voting to assign data points to final clusters. It then employs filter-based, wrapper-based, and embedded feature selection techniques within each cluster to identify common features. Additionally, feature selection is applied to the entire dataset to extract another set of common features. These selected features are combined to form the final best feature set. Experimental results demonstrate the efficacy of BUGOPTIMIZE compared to existing feature selection methods, reducing MAE and RMSE in Linear Regression (MAE: 0.2668 to 0.2609, RMSE: 0.3251 to 0.308) and Random Forest (MAE: 0.1626 to 0.1341, RMSE: 0.2363 to 0.224), highlighting its significant contribution to bug dataset optimization and prediction accuracy in software development while addressing feature selection limitations. By mitigating the disadvantages of current approaches and introducing a comprehensive and scalable solution, BUGOPTIMIZE presents a significant advancement in bug dataset optimization and prediction accuracy in software development environments
A Study on the Selection of Software Testing Techniques Based on the Multi-Criteria Decision-Making Model
Get PDFIn software development, software testing is a crucial activity that aims to ensure the quality and reliability of software products. However, the process of selecting the most appropriate software testing technique for a particular project can be a challenging and intricate task, as it involves multiple conflicting criteria and goals. This article suggests applying multi-criteria decision making (MCDM) methods, namely COPRAS, EDAS, and MABAC, to the challenge of selecting software testing techniques. Additionally, the weights of the criteria were analyzed using the MEREC method. The outcomes indicate that the methods employed consistently rank the options. The end-to-end testing technique is ranked the highest, while bottom-up integration testing is ranked the lowest. The ranking and selection approach proposed in the article can serve as a valuable tool for software testers and managers when making informed decisions about selecting software testing techniques that meet user requirements
Lstm Neural Networks and Iot Data for Predictive Maintenance in Healthcare
Get PDFThe most important in the modern provision of health care are medical devices that are involved in the process of prevention, diagnosis and treatment, rehabilitation. Ensuring their proper technical condition is the key to patient and user safety. However, the traditional ways of maintaining medical equipment are not enough for the increasing complexity of devices. By using information technology, social networking technologies, computerized systems digitization, and big data analytics, including machine learning, we have the ability to improve the quality of provision of services in the healthcare system. Predictive maintenance has become a fast-growing trend for assessing the technical condition of equipment and making predictions about possible failure scenarios to organize preventive maintenance. This systematic literature review will analyze previous research on predictive maintenance, with a special focus on its use in healthcare. The analysis of the articles found in several scientific search databases demonstrates that there is still much untapped potential for predictive maintenance in healthcare. This paper aims to introduce a new approach tuple, which will make it possible to provide proactive maintenance of medical equipment with the use of long short-term memory and Internet of things in healthcare analytics. This SLR will serve as a starting point to understand the predictive maintenance solutions in the industry, main findings, challenges, and new opportunities, and will give insights for future research regarding predictive maintenance
Alzheimer Detection System Using Hybrid Deep Convolutional Neural Network
Get PDFAlzheimer’s disease of the sixth leading causes of death in the United States of America is projected to grow to the third place of all causes of death for the elderly soon to cancer and heart decease. Timely detection and prevention are crucial to it. AD detection is based on multiple medical examinations which all lead to extensive multivariate heterogeneous data. This factor makes manual comparison, evaluation, and analysis hardly possible. The hereby study proposes a new approach to the detection of AD at the earliest stage hybrid deep learning algorithms. Several feature extraction and selection draw possible features. The method involves InceptionV3 and DenseNet for both pre-processing and classification tasks, while MobileNet enables data pre-processing and object detection. Experimental results with 100 epochs and 15 hidden layers show InceptionV3 has an accuracy of 98%, which outperforms other models available. The comparative analysis with other CNN models endorses the proposed method, achieving the highest performance across the board from our system
Recognition Character Sanskrit Using Convolution Neural Network
Get PDFThis research presents a pioneering approach using Convolutional Neural Networks (CNNs) for character recognition in Sanskrit, a language renowned for its intricate script and diverse character set. Addressing challenges posed by Sanskrit's complex script and historical variations in writing styles, we developed a CNN-based model that undergoes meticulous preprocessing to enhance image quality and normalize writing styles. Trained on a substantial dataset of annotated Sanskrit characters, our model showcases remarkable accuracy in recognizing Sanskrit characters, even amidst noise and diverse writing styles. This achievement holds significant implications for digitizing ancient manuscripts, aiding linguistic research, and preserving cultural heritage. Automating Sanskrit character recognition accelerates the analysis of Sanskrit texts, offering insights into linguistic evolution, cultural practices, and historical narratives. Moreover, this research lays a foundation for advancing character recognition techniques in complex scripts and languages, fostering opportunities for preserving and exploring diverse cultural heritages worldwide
Utilizing Random Forest for DDoS Attack Detection
Get PDFThe “Distributed Denial of Service” (DDoS) attack represents one of the most common forms of cyber assaults. Top of FormThe goal of DDoS is to overwhelm the server machine with an overwhelming number of data packets. This causes the bulk of the network bandwidth and server resources to be used leading to a Distributed denial-of-service problem. In this paper, we employed a random forest classifier for detecting the DDoS attack. This leads to an improvement in accuracy as well as a reduction in the amount of processing overhead required. Utilizing the CICDDOS2019 dataset, our experimental results showcased an impressive accuracy rate of 99.81%
Yoga and Artificial Intelligence: A Review of The Potential Applications of AI in Yoga Research and Practice for Neurological Disorders
Get PDFYoga has become an integral component of many people's lives around the world in recent years. Yoga emphasizes physical, mental, and spiritual links and is a health-promoting exercise method. On the other hand, doing yoga incorrectly and spraining your muscles can lead to pain and other health complications. A broad spectrum of issues impacting the nervous system are designated as neurological disorders. These conditions can be minor, like migraines, or severe, like multiple sclerosis and Parkinson's disease. Due to their complexity, many illnesses can be difficult to diagnose and treat. The use of holistic methods, especially yoga, has demonstrated potential in recent years for symptom relief and enhancing the general health of those with neurological illnesses. In addition, the development of artificial intelligence (AI) has improved our comprehension of these illnesses and offered creative approaches to individualised therapy regimens. The goal of artificial intelligence is to simulate, create, apply, and study the theory, technique, application system, and technology involved in increasing human intelligence. This paper discusses the benefits, drawbacks, and future prospects of applying artificial intelligence to the field of mental illness in an effort to serve as a resource for the field's sustainable development and also gives a succinct synopsis of the utilization of AI for examining yoga's impacts on neurological health and the potential for personalized therapie