International Journal of Information Technology

Block matching algorithms play vital role in the success of the video coding standards. A new block matching algorithm is proposed in this manuscript. Star diamond search with adaptive threshold is one of the state of the art algorithms for the motion estimation. Star diamond search with the adaptive threshold is enhanced by assigning the weights to the possible search points based on the chances to attain the matching block in various directions. This algorithm terminates early due to the cutoff threshold for the distortion at any stage of the algorithm. Algorithm also takes the advantage of the spatial coherence in adjacent blocks by assigning the highest precedence to the spatially left block. Proposed algorithm produces quality encoded frames with the significant improvement in the computation speed. Proposed algorithm achieves the speed gain in the range of 35–85%.

Self-care or personal care denotes those actions or doing that a person undertakes in supporting personal health, limiting personal illness, preventing personal disease and reinstating their own health. Self-caring is big challenge in exceptional/disabled children. With recent advancement in artificial intelligence in last few years, machine learning can be used for classification of self-care problem in children with different age groups. The paper proposed an enhanced expert system based on machine learning for diagnose and classification of self-care issues in children with physical and mental disorder. Partitioned Multifilter with Partial Swarm Optimization (PM-PSO) is used for attribute/feature selection and the outcomes are analogize with Principal Component Analysis (PCA). The preferred features/attributes are tested, trained and validated on following classifiers:-Naïve Bayes, Multilayer Perception (MLP), C-4.5 and Random Tree. tenfolded cross validation is used for validation, testing and training. PCA selects 32 attributes and shows truly categorized instances i.e. accuracy as: (1) 80% for Naïve Bayes; (2) 68.57% for MLP; (3) 68.57% for C 4.5 and; (4) 64.28% for Random Tree. The classifiers show a significant improvement in performance with PM-PSO feature selector. 50 attributes were selected with PM-PSO. It shows truly categorized instances/accuracy as: (1) 81% for Naïve Bayes; (2) 80% for MLP; (3) 80% for C 4.5 and; (4) 78.57% for Random tree.

Quantum-dot Cellular Automata (QCA) is an alternative step towards semiconductor-based CMOS technology. The QCA nanotechnology has a very fast circuit speed, ultra-low power consumption, minimum energy dissipation, extremely high switching frequency, and very small size in nanometer (nm) than traditional transistor-based CMOS technology. QCA technology plays an important role in the field of computational, communication, and information technology. In this paper, we have presented an experimental coplanar single layer reversible logic gates (RLG) such as Feynman gate (FG) and Double Feynman gate (F2G) implementation by using a bijective functional algorithm and demonstrated the zone partitioning problems, which are related to the synchronization of the clock. An optimized the energy dissipation of proposed circuit’s by QCADesigner-E (QD-E) version 2.2 tool. The presented reversible computational logic gates such as FG and F2G utilized 43.47% and 32.50% less number of design QCA cells, and 37.50% and 37.77% reduce the total design area respectively as compared to an existing optimal design. The proposed synchronization clocking method has outstanding characteristics such as reduced total design area, cell area, and latency, low complexity of circuits, and ultra-high-speed of proposed implemented designs.

Consumer-generated data provides a massive amount of market data that helps improve brands' decision-making processes within a highly demanding marketplace. This paper aims to investigate the dynamics behind Twitter user-generated content in relation to hijab/modest fashion based on a random sample of 144,800 tweets. Sentiment analysis was conducted, while a detection algorithm was implemented to identify the main influencers in relation to the hijab/modest fashion market. Results identify and profile the influencers and opinion leaders in the hijab/modest fashion global market. Results also show a high diversity of emojis usage in hijab-related tweets which highlighted the advantage of using them within hijab fashion brands’ communications. Finally, a partitioning around medoids (PAM) clustering method was applied to define consumer clusters. The clustering algorithm used highlights the heterogeneity and diversity of the global hijab fashion market. This study advances prior literature on hijab/modest-fashion consumers, and their opinions towards hijab brands. The study also helps marketers and decision-makers to understand consumer trends in this significant and emerging market.

Artificial Intelligence (AI) is increasingly being employed in critical decision-making processes such as medical diagnosis, credit approval, criminal justice, and many more. However, many AI models exploit complex algorithms that are difficult for humans to see through, which can lead to concerns about accountability, bias, and the ability to trust the outcomes. With the increasing demand for AI systems to be transparent, interpretable, and reliable, the field of Explainable AI (XAI) has gained attention of the researchers. This paper presents a robust hybrid architecture that combines Bidirectional Gated Recurrent Units (BiGRU) and Artificial Neural Networks (ANN) for the classification of texts and sentiment analysis. Interpretable Model Agnostic Explanation (LIME) has been employed with our proposed model to enhance confidence in the outcomes. The proposed architecture is found to be effective for sentiment analysis from texts, and classifying images containing handwrit- ten characters. It leverages the BiGRU to model the sequential dependencies in the data, while the ANN is used for the final classification. Evaluations on both Bengali and English datasets show that the proposed architecture outperforms state-of-the-art models in various performance metrics, providing meaningful and interpretable explanations for its predictions. The model can be used in systems that require the architectures to be computationally less demanding, yet a decent accuracy is secured.

Pattern Recognition is the method of mapping the inputs to their respective target classes based on features of data. In this paper a stacked ensemble meta-learning approach for customized convolutional neural network is proposed for Marathi handwritten numeral recognition. Stacked ensemble merges the pre-trained base pipe lines to create a multi-head meta-learning classifier that outputs the final target labels. It overpowers the average ensemble because the weighted and maximum contribution of each pipeline is taken in this approach. The stacked ensemble meta-learning classifier proves to be efficient because the base pipelines, which are already acquainted with output desirable results, are concatenated, instead of averaging, to achieve maximum efficiency. Performance evaluation and analysis have been done on Marathi handwritten numeral dataset, and the experiment results are better than the existing proposed systems.

Text is a rich and precise information source for understanding natural scene imagery and video. Text detection and localization improves the ability to understand text. Text detection and recognition faces numerous challenges such as noise, blur, distortion and variation. Though substantial research work has been done in the recent years, this area is still considered interesting by the research community and is open for improvements. Text detection and localization demands considerable large training datasets, computational ability and a prolonged training process. In this work, we have tried to address this issue by considering efficient algorithms to extract text contours and by transforming the image into a list of contour features. This list of contour features then be given to a convolutional neural network. We have shown that this would reduce training effort, ease text component classification and improve the results. After training with 300 images of MSRA-TD500, we are able attain a precision of 0.84, recall of 0.67 and a f-measure of 0.75. Also, contours are more effective than conventional rectangular bounding boxes in precisely localizing the text components.

In Japan, for tsunami disaster prevention, the government prepares tsunami hazard maps and provides them to residents. However, residents often do not examine the provided maps. Thus, these maps are not well utilized. In this study, we focused on the offline Geographic Information System and proposed a tsunami disaster drill support system that displays a tsunami hazard map, the current location, and route for the users. Using a prototype of the proposed system, we conducted an evaluation for the same in the Otaru City, Hokkaido. The results showed that the basic functions of the proposed system were mostly favorable for the evacuation process.

As cloud computing continues to grow, the energy consumption of cloud-edge resources has become a concern, particularly in terms of cost of energy and environmental effects. Therefore, reducing energy consumption in cloud-edge environments is an important issue that needs to be addressed to ensure sustainable and cost-effective cloud services. The existing approaches face challenges in achieving optimized energy consumption and workflow execution delay while maintaining reliability. Therefore, there is a need for a novel approach that can address these challenges and provide an effective solution for managing scientific workflows in a hybrid cloud environment. This paper introduces Resource Prediction and Scheduling Error Optimization (RPSEO), a novel approach for optimizing energy consumption and workflow execution delay in cloud-edge environments. The proposed method leverages a task-ordering web server management system and a soft-computing-based searching algorithm. Evaluation of Epigenomics and SIPHT workflows demonstrates significant improvements, surpassing existing methods Reliability-Aware Cost-Efficient Scientific (RACES), Delay Aware and Performance Efficient Energy Optimization (DAPPEO), and Reliable and Efficient Webserver Management (REWM) with better average energy consumption performance (up to 43.92% and 35.93% for Epigenomics and SIPHT) and cost efficiency (up to 44.53% and 73.50% for Epigenomics and SIPHT). RPSEO emerges as a promising solution for reliable and efficient scientific workflow management in hybrid cloud settings.