International Journal of Intelligent Systems and Applications @ijisa
Journal articles - International Journal of Intelligent Systems and Applications
All articles: 1264
Scientific article
This research aims to test the feasibility of Programmable Logic Controller implementation of an Artificial Neural Network based bearing fault diagnosis using vibration datasets. The main drawback of using a Programmable Logic Controller along with an Artificial Neural Network is that it does not support the parallel nature of neural networks. This drawback is not significant for relatively small applications like bearing diagnosis that involve very short execution time. In this paper, a three layer multilayer perceptron backpropagation neural network is trained using Levenberg-Marquardt training algorithm with vibration dataset consisting of four bearing status classes: normal, outer race way fault, inner race way fault and rolling element (ball) fault. Time-frequency domain and time domain input features were considered in this research. Both approaches have performed well during simulation phase. But the time-frequency feature extraction approach was observed to take too long scan cycle time to be implemented in real-time. This is due to the computationally intensive nature of Fast Fourier Transform algorithm involved during feature extraction. The time domain approach is proved to be feasible for Programmable Logic Controller implementation. The time domain input features used for neural network training were root mean square, variance, kurtosis and negative log likelihood values. The average performance obtained during simulation with 10-fold cross validation performance estimator was an error of 7.9 x10-4. The performance tests of Programmable Logic Controller implementation resulted in 100% bearing fault detection rate.
Free
A new approach for dynamic parametrization of ant system algorithms
Scientific article
This paper proposes a learning approach for dynamic parameterization of ant colony optimization algorithms. In fact, the specific optimal configuration for each optimization problem using these algorithms, whether at the level of preferences, the level of evaporation of the pheromone, or the number of ants, makes the dynamic approach an interested one. The new idea suggests the addition of a knowledge center shared by the colony members, combining the optimal evaluation of the configuration parameters proposed by the colony members during the experiments. This evaluation is based on qualitative criteria explained in detail in the article. Our approach indicates an evolution in the quality of the results over the course of the experiments and consequently the approval of the concept of machine learning.
Free
Scientific article
In this paper a new Quantum Tunneling Particle Swarm Optimization (QTPSO) algorithm is proposed and applied to the training of feedforward Artificial Neural Networks (ANNs). In the classical Particle Swarm Optimization (PSO) algorithm the value of the cost function at the location of the personal best solution found by each particle cannot increase. This can significantly reduce the explorative ability of the entire swarm. In this paper a new PSO algorithm in which the personal best solution of each particle is allowed to tunnel through hills in the cost function analogous to the Tunneling effect in Quantum Physics is proposed. In quantum tunneling a particle which has insufficient energy to cross a potential barrier can still cross the barrier with a small probability that exponentially decreases with the barrier length. The introduction of the quantum tunneling effect allows particles in the PSO algorithm to escape from local minima thereby increasing the explorative ability of the PSO algorithm and preventing premature convergence to local minima. The proposed algorithm significantly outperforms three state-of-the-art PSO variants on a majority of benchmark neural network training problems.
Free
A new robust and imperceptible image watermarking scheme based on hybrid transform and PSO
Scientific article
In this paper, a new robust and imperceptible digital image watermarking scheme that can overcome the limitation of traditional wavelet-based image watermarking schemes is proposed using hybrid transforms viz. Lifting wavelet transform (LWT), discrete cosine transform (DCT) and singular value decomposition (SVD). The scheme uses canny edge detector to select blocks with higher edge pixels. Two reference sub-images, which are used as the point of reference for watermark embedding and extraction, have been formed from selected blocks based on the number of edges. To achieve a better trade-off between imperceptibility and robustness, multiple scaling factors (MSF) have been employed to modulate different ranges of singular value coefficients during watermark embedding process. Particle swarm optimization (PSO) algorithm has been adopted to obtain optimized MSF. The performance of the proposed scheme has been assessed under different conditions and the experimental results, which are obtained from computer simulation, verifies that the proposed scheme achieves enhanced robustness against various attacks performed. Moreover, the performance of the proposed scheme is compared with the other existing schemes and the results of comparison confirm that our proposed scheme outperforms previous existing schemes in terms of robustness and imperceptibility.
Free
A novel approach for regression testing of web applications
Scientific article
Software testing is one of the most arduous and challenging phase which is to be implemented with the intention of finding faults with the execution of minimum number of test cases to increase the overall quality of the product at the time of delivery or during maintenance phase. With the ever increasing demand of web applications and to meet never ending customer expectations, updations are to incorporate which will be validated through testing process. The structure of the web applications (dynamic website) can be modeled using weighted directed graph which consists of numerous paths starting from homepage (index page) of the website. For thorough testing of the website each and every path of the graph should be tested but due to various constraints like time, money and human resources it becomes very much impractical. This scenario ultimately gives rise to the motivation for the development of technique which reduces the number of paths to be tested so that tester community can test only these numbers of path instead of all possible paths so that satisfactory number of faults can be exposed. In this proposed approach assignment of weights on the edges of the directed graph takes place on the basis of the organization of the website, changes in the structure of the website at page level, experience of the coder and the behaviour of the users who have visited the website earlier. The most fault prone paths are identified using random, greedy, Ant Colony Optimization (ACO) and Artificial Bee Colony Optimization (ABCO) algorithms. Two small size websites and one company’s website, and their two versions, were considered for experimentation. Results obtained through ACO and ABCO are promising in nature. This approach will support testing process to be completed in time and delivery of the updated version within given hard deadlines.
Free
A novel evolutionary automatic data clustering algorithm using teaching-learning-based optimization
Scientific article
Teaching-Learning-Based Optimization (TLBO) is a contemporary algorithm being used as a novel, trustworthy, precise and robust optimization technique for global optimization over continuous spaces both constrained and unconstrained tribulations. TLBO works on the beliefs of teaching and learning and clearly justifies this pedagogy by highlighting the effect of power of a teacher on the output of learners in a class. This paper, explores the applicability of k-means unsupervised learning into TLBO with two endeavors, i.e. to automatically find the optimal number of naturally classified partition in the data without any prior information, and the other is to inspect the naturally classified partitions with cluster validity indices (CVIs) and endorse the goodness of clusters. The proposed automatic clustering algorithm using TLBO (AutoTLBO) pursues a novel evolutionary approach by incorporating the simple k-means algorithm and CVIs into TLBO to configure and validate automatic natural partition in datasets. This algorithm retains the core ideology of clustering to minimize the inter cluster distances and maximize the intra cluster distances among the data. Experimental analysis substantiates the openness of the anticipated method after inspecting suavest panoramic rendering over artificial and benchmark datasets.
Free
A novel handoff necessity estimation approach based on travelling distance
Scientific article
Mobility management is one of the most important challenges in Next Generation Wireless Networks (NGWNs) as it enables users to move across geographic boundaries of wireless networks. Nowadays, mobile communications has heterogeneous wireless networks offering variable coverage and Quality of Service (QoS). The availability of alternatives generates a problem of occurrence of unnecessary handoff that results in wastage of network resources. To avoid this, an efficient algorithm needs to be developed to minimize the unnecessary handoffs. Conventionally, whenever a Wireless Local Area Network (WLAN) connectivity is available, the mobile node switch from the cellular network to wireless local area network to gain maximum use of high bandwidth and low cost of wireless local area network as much as possible. But to maintain call quality and minimum number of call failure, a considerable proportion of these handovers should be determined. Our algorithm makes the handoff to wireless local area network only when the Predicted Received Signal Strength (PRSS) falls below a threshold value and travelling distance inside the wireless local area networkis larger than a threshold distance. Through MATLAB simulation, we show that our algorithm minimizes the probability of unnecessary handoff, and probability of handoff failure. Hence, the proposed algorithm is able to improve handover performance.
Free
A novel intelligent ARX-Laguerre distillation column estimation technique
Scientific article
In practical applications, modeling of real systems with unknown parameters such as distillation columns are typically complex. To address issues with distillation column estimation, the system is identified by a proposed intelligent, auto-regressive, exogenous-Laguerre (AI-ARX-Laguerre) technique. In this method, an intelligent technique is introduced for data-driven identification of the distillation column. The Laguerre method is used for the removal of input/output noise and decreases the system complexity. The fuzzy logic method is proposed to reduce the system’s estimation error and to accurately optimize the ARX-Laguerre parameters. The proposed method outperforms the ARX and ARX-Laguerre technique by achieving average estimation accuracy improvements of 16% and 9%, respectively.
Free
A novel text representation model to categorize text documents using convolution neural network
Scientific article
This paper presents a novel text representation model called Convolution Term Model (CTM) for effective text categorization. In the process of text categorization, representation plays a very primary role. The proposed CTM is based on Convolution Neural Network (CNN). The main advantage of proposed text representation model is that, it preserves semantic relationship and minimizes the feature extraction burden. In proposed model, initially convolution filter is applied on word embedding matrix. Since, the resultant CTM matrix is higher dimension, feature selection methods are applied to reduce the CTM feature space. Further, selected CTM features are fed into classifier to categorize text document. To discover the effectiveness of the proposed model, extensive experimentations are carried out on four standard benchmark datasets viz., 20-NewsGroups, Reuter-21758, Vehicle Wikipedia and 4 University datasets using five different classifiers. Accuracy is used to assess the performance of classifiers. The proposed model shows impressive results with all classifiers.
Free
Scientific article
The numbers of educational institutions are growing at par with the lost student rate in a country like India. When a missing student is found we need to identify the student on the strength of some common parameter like student name, his/her institution name, branch or class etc. But we never get accurate and complete information in most of the cases to identify or recognize a lost student. In such a situation, a soft computing model can be a striking choice to track a lost student on the basis of partial information. In the past we propose soft computing model for the same. This paper proposes a more optimized parallel soft computing model which takes half of the time taken by the earlier single thread model for identifying a lost student on the basis of imprecise and partial information. The system is tested meticulously on a database of 50000 records and an efficiency of 94% is obtained.
Free
Scientific article
Due to the global demand for energy saving and reduction of greenhouse gas emissions, utilization of renewable energy sources have increased in electricity networks. The negative aspects of this technology are very complex and not well known which affect reliability and robustness of the grids. Microgrids based on renewable energy sources have gained significant popularity, due to the major benefits it has to offer for solving the increasing energy demand. Harmonic distortion in microgrids caused by the non-linear loads is an essential topic of study necessary for the better understanding of power quality impacts in microgrids. The various control techniques utilized to curtail the power quality impacts on micro grids are reviewed in this paper. Also, Optimization based control techniques utilized for power quality improvement in microgrids is discussed in this review.
Free
A study on liver disease diagnosis based on assessing the importance of attributes
Scientific article
Liver is a needful body organ that forms an important barrier between the gastrointestinal blood, which contains large amounts of toxins, and antigens. Liver diseases contain hepatitis B and hepatitis C virus infections, alcoholic liver disease, nonalcoholic fatty liver disease and associated cirrhosis, liver failure and hepatocellular carcinoma are primary causes of death. The main purpose of this study is to investigate which attributes are important for effective diagnosis of liver disorders by performing the machine learning approach based on the combination of Stability Selection and Random Forest methods. In order to generate more accuracy, dataset was balanced by utilizing the Random Under-Sampling method. Important ones in all attributes were detected by utilizing the Stability Selection method which was performed on sub-datasets, which were obtained with 5 fold cross-validation technique. By sending these datasets to the Random Forest algorithm, the performance of the proposed approach was evaluated within the frame of accuracy and sensitive metrics. The experimental results clearly show that the Random Under-Sampling method can potentially improve the performance of the combination of Stability Selection and Random Forest methods in machine learning. And, the combination of these methods provides new perspectives for the diagnosis of this disease and other medical diseases.
Free
A swarm intelligence based chaotic morphological approach for software development cost estimation
Scientific article
In the last century, with the inception of various software development industries at around mid-1960’s, the complexities and size of the software have always been a major concern for the industries. The ad-hoc process of development has evolved into a standardized one due to the increase in the size and complexity of software projects. The standardized process of software development was further evolved to predict the overall cost required for the development before the software is actually built. To achieve the same, many cost estimation methodologies have already been successfully implemented, each with certain pros and cons. The present scenario demands even further refined and accurate predictions, which the above-said methods cease to provide. In this paper, we present a chaotically modified particle swarm optimization (CMPSO) based morphological learning approach to accurately estimate the cost incurred in the development process. The proposed approach focuses on a mathematical morphological (MM) framework based hybrid artificial neuron (also called dilation-erosion perceptron or DEP) with algebraic foundations in complete lattice theory (CLT). The proposed CMPSO-DEP model was tested on 5 well-known datasets of software projects with three popular performance metrics and the results were compared with the best existing models available in the literature.
Free
A trend analysis of machine learning research with topic models and Mann-Kendall test
Scientific article
This paper aims to systematically examine the literature of machine learning for the period of 1968~2017 to identify and analyze the research trends. A list of journals from well-established publishers ScienceDirect, Springer, JMLR, IEEE (approximately 23,365 journal articles) related to machine learning is used to prepare a content collection. To the best of our information, it is the first effort to comprehend the trend analysis in machine learning research with topic models: Latent Semantic Analysis (LSA), Latent Dirichlet Allocation (LDA), and LDA with Coherent Model (LDA_CM). The LDA_CM topic model gives the highest topic coherence amongst all topic models under consideration. This study provides a scientific ground that helps to overcome the subjectivity of collective opinion. The Mann-Kendall test is used to understand the trend of the topics. Our findings provide indicative of paradigmatic shifts in research methodology of significant patterns of topical prominence and the evolving research areas. It is used to highlight the evolution regarding the previous and recent trends in research topics in the area of machine learning. Understanding such an intellectual structure and future trends will assist the researchers to adopt the divergent developments of this research in one place. This paper analyzes the overall trends of the machine learning research since 1968, based on the latent topics identified in the period of 2007~2017 that may be helpful to the researchers exploring the recommended areas and publish their research articles.
Free
ABC Algorithm Based Interval Type-2 Fuzzy Logic Controller for an Inverted Pendulum
Scientific article
In this paper, a hybrid control technique is proposed for managing the variation of angle and velocity of the inverted pendulum. The proposed hybrid technique is the combination of ABC algorithm and interval type-2 Fuzzy Logic System (IT2FLS). The objective of the proposed hybrid control technique is to achieve the stability position of the pendulum. Here, the ABC algorithm is used to optimize the change of angle and change of velocity of the pendulum. With the optimized value, the optimal membership functions and the interference system are developed using IT2FLS. Using the ABC based IT2FLS, the position of the inverted pendulum is maintained towards the reference position. The proposed hybrid control technique is implemented in MATLAB/Simulink working platform and the control performances are evaluated.
Free
AI Scheduling with Contextual Transformers
Scientific article
This paper presents a comprehensive framework for intelligent and personalized task scheduling based on Transformer architectures and contextual-behavioral feature modeling. The proposed system processes sequences of user activity enriched with temporal, spatial, and behavioral information to generate structured task representations. Each predicted task includes six key attributes: task type, execution time window, estimated duration, execution context, confidence score, and priority level. By leveraging Transformer encoders, the model effectively captures long-range temporal dependencies while enabling parallel processing, which significantly improves both scalability and responsiveness compared to recurrent approaches. The system is designed to support real-time adaptation by integrating diverse data sources such as device activity, location, calendar status, and behavioral metrics. A modular architecture enables input encoding, multi-head self-attention, and global behavior summarization for downstream task generation. Experimental evaluation using artificially generated user data illustrates the model’s ability to maintain high accuracy in task type and timing prediction, with consistent performance under varying contextual conditions. The proposed approach is applicable in domains such as digital productivity, cognitive workload balancing, and proactive time management, where adaptive and interpretable planning is essential.
Free
AI-driven Innovations for Sustainable Hospital Waste Management in Developing Countries
Scientific article
Population growth and pandemics like COVID-19 have led to the depletion of natural resources and an increase in hospital waste generation. This issue is particularly pressing in developing countries, where innovative solutions are needed to address the environmental and health risks associated with improper waste disposal. Traditional waste sorting methods, which rely on human intervention, are time-consuming and pose a significant risk of infection. Moreover, different categories of hospital waste require specific treatment methods. This study proposes an artificial intelligence-based approach for classifying and sorting hospital waste using Convolutional Neural Networks (CNNs). The proposed CNN model effectively identifies and categorizes various types of hospital waste, providing a sustainable solution that enhances regulatory compliance. The model is developed using an approach which leverages K-fold cross-validation, data augmentation, and a publicly available, modest-sized dataset tailored for variability in waste categories. The model achieved a peak classification accuracy of 97.08%, along with strong precision, recall and F1-score, despite class imbalances and the presence of visually similar and hard-to-distinguish waste categories. These results highlight the potential of the model to improve hospital waste management practices, thereby reducing environmental impact and health risks.
Free
AI-powered Predictive Model for Stroke and Diabetes Diagnostic
Scientific article
Research efforts in the prediction of stroke and diabetes prioritize early detection in order to enhance patient outcomes. To achieve this, a variety of methodologies are integrated. Existing studies, on the other hand, are marred by imbalanced datasets, lack of diversity in their datasets, potential bias, and inadequate model comparisons; these flaws underscore the necessity for more comprehensive and inclusive research methodologies. This paper provides a thorough assessment of machine learning algorithms in the context of early detection and diagnosis of stroke and diabetes. The research employed widely used algorithms, including Logistic Regression, Decision Tree, Random Forest, K-Nearest Neighbors (KNN), and XGBoost Classifier, to examine medical data and derive significant findings. The XGBoost Classifier demonstrated superior performance, with an outstanding accuracy, precision, recall, and F1-score of 87.5%. The comparative examination of the algorithms indicated that the Decision Tree, Random Forest, and XGBoost classifiers consistently exhibited strong performance across all measures. The models demonstrated impressive discrimination capabilities, with the XGBoost Classifier and Random Forest reaching accuracy rates of roughly 87.5% and 86.5% respectively. The Decision Tree Classifier exhibited notable performance, with an accuracy rate of 83%. The overall accuracy of the models was evident in the F1-score, a metric that incorporates recall and precision, where the XGBoost model exhibited a marginal improvement of 2% over the Random Forest and Decision Tree models, and 4.25 percent over the last two. The aforementioned results underscore the effectiveness of the XGBoost Classifier, which will be employed as a predictive model in this study, alongside the Random Forest and Decision Tree models, for the accurate identification of stroke and diabetes. Furthermore, combining datasets improves model performance by utilizing relative features. This integrated dataset improves the model's efficiency and creates a resilient and comprehensive prediction model, improving healthcare outcomes. The findings of this research make a valuable contribution to the advancement of AI-driven diagnostic systems, hence enhancing the quality of healthcare decision-making.
Free
AI’s Current Impact and Future Potential in Emergency Services: A Comprehensive Review and Analysis
Scientific article
The dynamic force of artificial intelligence (AI) is reshaping our world, not in the distant future, but today. Its transformative potential, adaptability, and capacity to liberate human potential are becoming evident in a multitude of domains. AI's ability to process vast datasets, offer data-driven recommendations, and enhance decision-making processes underscores its pivotal role in addressing complex challenges. This article explores AI's current impact and its potential for further growth. It reviews 77 articles across diverse domains, highlighting AI's role in emergency services. Through an in-depth analysis of these studies, the paper provides a broad overview of the current state of AI in emergency services, identifying key trends, challenges, and future opportunities. By examining the methodologies, datasets, AI and deep learning techniques, feature selection processes, evaluation metrics, and prediction models used in each study, the paper aims to offer a thorough understanding of AI's role in this critical sector. This extensive body of knowledge is intended to be a valuable resource for researchers, practitioners, and policymakers. It supports the ongoing advancement of AI-driven emergency services, with the goal of saving lives, optimizing resource allocation, and enhancing response times in critical situations. Ultimately, this collaborative effort seeks to foster the development of more resilient and responsive emergency systems that can effectively mitigate risks and deliver timely aid to those in need. By advancing the capabilities of emergency response systems, AI enhances the precision and efficiency of critical interventions, ultimately leading to better outcomes and improved resilience in crisis situations.
Free
ANN model of border regions development: approach of closed systems
Scientific article
In this article we have suggested a new method of regional systems study that is based on model environmental influences isolation on their parameters dynamics. The presented model deepens greatly the investigation process of complex systems and allows defining clearly its functioning peculiarities without significant reduction in number of system characteristics as if we have simple technical or physical objects of knowledge. The described method, together with the statistical control data, is used for other social and economic objects research. The successful model testing in the form of artificial neural network model of Chernivtsi region static parameters has revealed the peculiarities of its interaction with European neighbors. In particular, for the first time we have defined their contribution to the increase of some social and economic indices on the period 2005-2015, that cannot be explained by other methods, such as correlation and regressive analysis. Applied use of the isolated investigation idea of the complex meso level systems together with the technology of data mining allows solving many actual tasks nominated by the regional administration practical workers.
Free
