Novel vision transformer and data augmentation technique for efficient detection of monkeypox disease
Multimedia Tools and Applications · مجلد 84 (27) · 31955-31973
Research outputs
مقالات ومراجعات وأوراق مؤتمرات وفصول كتب، مفهرسة برقم DOI ومربوطة بباحثي الجامعة.
5 مُخرَج نتائج البحث عن «multi-class classification»
Multimedia Tools and Applications · مجلد 84 (27) · 31955-31973
IEEE International Conference on Innovations in High Speed Communication and Signal Processing (IHCSP) · 1-8
The proposed IoT intrusion detection model uses machine learning to compensate for data loss in convolutional neural networks' initial pooling layer and the gradient vanishing problem that recurrent neural networks face when processing long sequential input. Time series data for intrusion detection models is optimized using Gramme matrix field visualization. To reduce data loss, the SoftPool pooling layer replaced the original. The recurrent neural network uses a bidirectional gated recurrent unit with fewer parameters and a multi-head structure modeled after the self-attention mechanism. Baseline data will be TON-IoT data. The number of heads multi-head bidirectional gated recurrent unit, pooling layer topologies, and visualization approaches are experimental parameters. The simulation results show that a three-head structure, SoftPool for layer pooling, and for accuracy, precision, detection rate, F1 score, and false positive rate, multi-class classification works well with field and Gramme matrix visualization. The suggested model outperforms state-of-the-art models in accuracy, detection rate, F1 score, and false positive rate by 0.14, 0.09, 0.13, and 0.08. Modern machine learning improves IoT intrusion detection.
International Conference on Inventive Computation Technologies (ICICT) · 112-117
Support vector machine has emerged as one of the most crucial classification techniques in recent years in pattern recognition, machine learning, and data mining. The model training will become more challenging when dealing with multi-classification problems, and its training time will increase considerably as the number of samples increases. A quick training data reduction technique MOIS that works with multiple classification problems is provided to address the issues above. This method takes the cluster centre as the reference point. While removing redundant training samples, selects the decisive boundary samples to significantly reduce the training data and reduce the distribution imbalance between categories. Experimental results show that MOIS can maintain or even improve the classification effect of support vector machines. Moreover, it can significantly improve training efficiency. For example, on the Opt digit data set, the proposed method improves the classification accuracy from 98.94% to 99.05% and, at the same time, shortens the training time to 15%; On the data set composed of the first 100 categories, when the accuracy rate is slightly improved (from 99.29% to 99.30%), the training time is significantly shortened to less than 6% of the original. In addition, MOIS itself has High operating efficiency.
International Conference on Emerging Research in Computational Science (ICERCS) · 1-6
Knee Osteoarthritis (KOA) is a common musculoskeletal condition that has a significant worldwide influence on patient well-being. Despite its importance, manually detecting KOA poses tremendous obstacles. Traditional diagnostic methodologies heavily depend on the visual interpretation of radiographic images by certified radiologists. This approach leads to laborious assessments and introduces the potential for discrepancies among observers. Effective manual identification is further complicated by the intricate structure of the KOA grading system, which evaluates joint space narrowing, osteophyte production, and overall degeneration. This paper presents a method for addressing these challenges by integrating the capabilities of Deep Learning (DL) and Machine Learning (ML) models. The dataset utilized in this research was acquired from Kaggle and comprises image samples that have been assigned KOA grades between 0 to 4. The data source is utilized to generate two distinct datasets. The initial dataset is structured as a binary classification task, wherein images are categorized as 'KOA' (grades 2 to 4) or 'healthy' (grades 0 and 1). The second dataset is the same as the original dataset, with five classes for each KOA grade. An important objective of this research is to assess the performance of the model in both binary and multi-class classification. The datasets go through pre-processing phases such as segmentation and normalization before being fed to a CNN (Convolutional Neural Network) model to extract significant features. Following this, the gathered features are utilized to train and validate numerous ML models. The assessment is predicated on several critical metrics that give information about the effectiveness of the model. The results of our study indicate that the Support Vector Machine (SVM) demonstrates outstanding efficiency in the area of binary classification. In contrast, the Random Forest (RF) model performs exceptionally well in multi-class classification, yielding promising outcomes. The suggested method not only addresses the complications of KOA grading, but it also presents a viable technique for early diagnosis and severity categorization, hence assisting clinicians and lowering the diagnostic burden on primary radiologists.
Lecture notes in networks and systems · 413-423