International Journal for Global Academic & Scientific Research
Not a member yet
64 research outputs found
Sort by
Leveraging AI for Accurate Time Series Forecasting
Full text linkThis study seeks to develop a robust model for forecasting time series data, with an eye towards complex temporal datasets. Accurate forecasting in time series analysis is a function of past information and constitutes a basis for unsupervised machine learning. With deep learning techniques such as neural networks, this work seeks to provide high accuracy over traditional approaches in time series forecasting. Such complex techniques have a significant impact in overcoming complications in forecasting in areas such as weather trends, consumption of energy, and financial trends in the marketplace. Out of such techniques, Artificial Neural Networks have been seen to outshine alternatives such as Long Short-Term Memory networks in working with complex temporal relationships. In this work, an opportunity for leveraging complex AI techniques towards enhancing accuracy and dependability in forecasting in a time series is focused on
A Hybrid Deep Learning Model for Classification of Psoriasis Disease
No full textPsoriasis is a chronic dermatological condition that manifests in multiple clinical forms, making its diagnosis and classification challenging. In this study, we present a hybrid deep learning framework that combines EfficientNet and Bidirectional Long Short-Term Memory (Bi-LSTM) networks for computerized psoriasis category. The dataset comprises of seven classes, Erythrodermic, Pustular, Plaque, Nail, Inversus, Guttate, and Healthy, containing a total of 1,420 dermoscopic images, divided into 994 images for training, 213 for validation, and 213 for testing following a 70:15:15 ratio. The images had been pre-processed and segmented the usage of a Snake-primarily based set of rules earlier than function extraction. To compare the effectiveness of the proposed model, we as compared it towards two different hybrid architectures (VGG16+LSTM and AlexNet+LSTM). Experimental results demonstrate that the EfficientNet+Bi-LSTM model achieves an overall accuracy of 86%, outperforming VGG16+LSTM (66%) and AlexNet+LSTM (68%). Beyond accuracy, the proposed model also showed improved macro-F1 and balanced per-class recall, indicating robustness in handling class imbalance. These findings suggest that integrating EfficientNet’s powerful feature extraction with Bi-LSTM’s sequential learning capability provides a promising direction for psoriasis disease classification
Enhanced Machine Learning Model for CVD Prediction Using Principal Component Analysis (PCA)
Full text linkThe World Health Organization (WHO) report says that each year, cardiovascular diseases are the leading reason for around 17.9 million deaths across the globe. This is a more serious problem in low- and middle-income countries where there are barriers to early check-ups and specific treatments. The quicker and better detection of heart attacks helps reduce the risk of death. Based on previous methods, the study takes the Cleveland Heart Disease dataset from the UCI Machine Learning repository and uses it to design and check best machine learning models that take advantage of standardization, Principal Component Analysis (PCA) and hyperparameter tuning. We used machine learning algorithms such as Support Vector Machine, k-Nearest Neighbors, Logistic Regression and a Multi Layer Perceptron model, all combined under a Voting Classifier. With a 98.33% accuracy, 98.25% F1-score, 96.55% precision, and 100% recall on test data, the enhanced hybrid model (Voting Classifier) leaves all other models far behind in performance. The hybrid model had small gaps between train and test values for metrics, with 1.24% accuracy difference, 1.29% F1-score difference, 3.45% precision difference and -0.92% recall difference. Incorporating Principal Component Analysis (PCA) lowered the number of dimensions used while increasing accuracy, precision and F1 scores for a number of models. The results suggest that the use of Principal Component Analysis (PCA)-combined hybrid models leads to better, more understandable and trustworthy tools for predicting CVD. Strengthening predictive models for CVD risk assessment is now possible, supporting prompt clinical choices and helping patients improve
Sustainable Truck Overload Management Framework
Full text linkThe Sensor for Overloading of Trucks project seeks to develop an advanced sensor mechanism with a high accuracy for checking whether a truck is overloaded or not. Eliminating overloading in trucks is critical for effective loading and weighing, reducing mechanical failure, minimizing deterioration in roads, and enhancing overall security policies in terms of roads. Overloading is one of the most important factors in causing accidents, infrastructure deterioration, and increased maintenance, and its management is a matter of high concern. The system developed in this work utilizes strain sensors for monitoring the compressive and tensional loads experienced at specific parts of a truck at which most strain is encountered. Measuring such a process, nevertheless, proves to be a challenge with a moving truck, whose motion generates variable and unpredictable jerks and rough roads, and temporarily generates fluctuations in strain, creating a problem in taking proper readings. The work seeks to overcome such complications through a robust and effective model of a sensor capable of working under such variable motion and providing proper readings for weighing and supporting safer transportation processes
Blockchain Technology for Supply Chain Management: Enhancing Transparency and Efficiency
Full text linkThe innovation of blockchain technologies into supply chain management is incremental by nature, attending to waste aspects and issues that are very old within SCM. As such, this reflection seeks to highlight the multifaceted benefits of blockchain in enhancing simplicity, traceability, security, and efficiency within supply chains. Blockchain provides a decentralized, permanent system of record that allows trust and honesty between partners; all valuable information is recorded in an imperishable way. Moreover, intelligent contracts mechanize some forms involved in supply chains, such as the processing of payments and inventory management, which reduces to a large extent not only regulatory costs but also human error. The high level of security measures is ensured by blockchain’s ability to encrypt this data and thus ensure information availability in case of sophisticated cyber threats. Such improvement in security reduces extortion and tampering, providing a more secure supply chain. This innovation further simplifies the administrative compliance, providing an accessible, verifiable record of each exchange. Not only this, but it also offers a way to verify whether material sources and technical support standards meet prevailing standards, thereby promoting brand identity to customers for the brand marked on a product and enhancing customer trust in a brand
Data-Driven Management: The Impact of Big Data Analytics on Organizational Performance
Full text linkInevitably, in such a fast-moving landscape of big data analytics lies the transformative opportunity for any organization to unlock new avenues of growth, operational efficiency, and strategic decision-making. This paper contributes a comprehensive methodology that will help businesses take advantage of big data analytics to secure a continued competitive advantage. At the core of this methodology is to have a robust data governance framework that will establish the security, integrity, and accessibility of the enterprise data assets by specifying relevant policies, processes, and technologies. This, in turn, within such a framework, allows state-of-the-art AI-driven anomaly detection mechanisms for encryption and access control to implement protection measures around sensitive information while enabling secure yet efficient data utilization. The methodology also continues its approach of putting in place an integrated data ecosystem that brings together different pockets or sources of data, such as real-time operation data, customer interaction, and unstructured information. A critical component of this methodology is putting in place the advanced predictive analytics capabilities that could be run based on tapping into the power of machine learning algorithms; by doing so, the organization will be predicting market trends, customer behavior, and risks highly accurately. This will be a very proactive way of making decisions that would let the business efficiently use the available resources, innovate products and services ahead of time, and create a distinctly competitive advantage. The transformative ability of this methodology for big data analytics opened new channels toward growth and innovation and firmly established the organization as an industry leader with long-lasting competitive advantage. The place of data-driven insight, data culture, and responsible data practice has been the key to success in this organization
To Study the Effect of Job Satisfaction on the Performance of Academic Faculties Working in Private Colleges and Private Universities in Indore
Full text linkThe objective of the existing study was to examine the impact of job satisfaction on the performance of employees working in private colleges and universities in Indore, India. To achieve this, questionnaires were distributed to a total of 60 employees, out of which 54 responses were received and considered as the sample from private colleges and universities in Indore. An equal number of employees (n = 54) were randomly selected from different types of organizations, including undergraduate and postgraduate colleges, as a comparison group. The study utilized a self-constructed questionnaire based on the Minnesota Satisfaction Questionnaire (MSQ-quick form) developed by Weiss et al. (1967), as well as a self-constructed Performance Evaluation Form (PRF). Initially, the reliability of both instruments was assessed to determine the significance of the scales. The study findings indicated a significant correlation between the type of occupation and job satisfaction. Moreover, a positive relationship between job satisfaction and employee performance was also observed. Therefore, the study concluded that satisfied employees performed better compared to dissatisfied employees, thus playing a significant role in the advancement of their organizations. Consequently, it is crucial for every organization to adopt specific strategies and methods to motivate and ensure employee satisfaction, thereby promoting high performance
Antimicrobial Properties Of Pyrazolone Compounds With Their QSAR Investigations
Full text linkThe present paper reports the analysis and the antimicrobial activity viz. antifungal activity of some pyrazolone compounds which were performed against Alternaria solani using disc diffusion method on nutrient agar and nutrient broth media. The compounds were characterized by elemental analysis and spectral studies. Result of antimicrobial screening indicated that compounds 4-Amino antipyrine thiosemicarbazone, 1-(2-Chloro-5-sulfophenyl)-3-methyl-5-pyrazolone and 1-(4 -Sulfoamidophenyl)-3-methyl-5-pyrazolone were found to be the most potent/active against A. solani. Correlation studies between different computed properties of the compounds with their biological activity, QSAR (Quantitative Structure Activity Relationship) investigations with a stepwise linear regression analysis were conducted. The studies are carried out using Hyperchem 8.0 version software using AM1, PM3, MNDO and ZINDO methods. Selected QSAR/ 3D-QSAR equations including different physical parameters of these series are reported
Social Media as a Tool for Cultural Preservation among Diaspora Communities
Full text linkSocial media has become a vital tool for diaspora groups to preserve their cultural legacy in an globally connected world. This study examines how distributed communities preserve and spread their cultural customs, languages, and practices through the use of digital media. It looks at how diaspora communities use social media to maintain their cultural identities and cultural heritage via particular case studies. The study highlights how social media may effectively promote cross-cultural communication, facilitate in the development of communities, and help close the gap between the old and new worlds. The study\u27s findings provide understanding on the dynamics of cultural preservation in the digital era and the ways in which social media might promote a variety of cultural environments throughout the world
Analysis of Single Server Markovian Queueing Model with Differentiated Working Vacation, Vacation Interruption, Soft Failure, Reneging of Customers
Full text linkThis article investigates the reneging of customers in M/M/1 model with differentiated working vacation, vacation interruption and soft failure. The customers come with rate λ and receives service during busy period with rate μ, where λ and μ obeys markovian distribution. In this model two distinct vacations are considered: one has been taken just after serving all customers in busy period with slow service rate θ as some soft failure occurs during working vacation (Vacation I). At an epoch of completion of working vacation, if any customers are present in the system, then the server moves to busy period for serving customers otherwise move to vacation II. During vacation II if customer comes then interruption is assumed to occur in the vacation and server returns to busy period otherwise remain in vacation .When an arriving customer finds server is on working vacation, it makes customer impatient and it start up an impatient timer T0 with an exponentially distributed rate α0 If service does not begin before T0 expires, the customer might renege with probability p without getting served or wait for their turn with probability 1-p=q. By using PGF technique we have derived different steady state probabilities and various system performances analytically. Effect of few parameters on different system performances have been shown numerically and illustrated graphically.