Browse Items (11855 total)
Sort by:
-
Classification Framework for Fraud Detection Using Hidden Markov Model
Machine learning is described as a computer program that learns from experience E with regard to some task T and some performance measure P, if its performance on T improves with E as measured by P. Suppose we have a credit card fraud detection which watches which transactions we mark as fraud or not, and on the basis, it knows how to filter better fraudulent transactions then, E is watching your transactions is fraud or not, T is classifying your transactions as fraud or not, P is number of transactions correctly differentiated as spam or not spam. Machine learning has two types: supervised learning and unsupervised learning. Supervised learning is the type of machine learning where machine is provided with input mapped with its output, and these inputs and outputs are used to make a machine learn a particular function from the trained dataset. There are two branches of supervised learning, i.e., classification and regression. In unsupervised learning, we do not supervise model instead we allow machine to work on its own to discover information. Clustering is type of unsupervised learning. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Classification of a New-Born Infant's Jaundice Symptoms Using a Binary Spring Search Algorithm with Machine Learning
A yellowing of the skin and eyes, called jaundice, is the consequence of an abnormally high bilirubin concentration in the blood. All across the world, both newborns and adults are afflicted by this illness. Jaundice is common in new-borns because their undeveloped livers have an imbalanced metabolic rate. Kernicterus is caused by a delay in detecting jaundice in a newborn, which can lead to other complications. The degree to which a newborn is affected by jaundice depends in large part on the mitotic count. Nonetheless, a promising tool is early diagnosis using AI-based applications. It is straightforward to implement, does not require any special skills, and comes at a minimal cost. The demand for AI in healthcare has led to the realisation that it may have practical applications in the medical industry. Using a deep learning algorithm, we created a method to categorise jaundice cases. In this study, we suggest using the binary spring search procedure (BSSA) to identify features and the XGBoost classifier to grade histopathology images automatically for mitotic activity. This investigation employs real-time and benchmark datasets, in addition to targeted methods, for identifying jaundice in infants. Evidence suggests that feature quality can have a negative effect on classification accuracy. Furthermore, a bottleneck in classification performance may emerge from compressing the classification approach for unique key attributes. Therefore, it is necessary to discover relevant features to use in classifier training. This can be achieved by integrating a feature selection strategy with a classification classical. Important findings from this study included the use of image processing methods in predicting neonatal hyperbilirubinemia. Image processing involves converting photos from analogue to digital form in order to edit them. Medical image processing aims to acquire data that can be used in the detection, diagnosis, monitoring, and treatment of disease. Newburn jaundice detection accuracy can be verified using image datasets. As opposed to more traditional methods, it produces more precise, timely, and cost-effective outcomes. Common performance metrics such as accuracy, sensitivity, and specificity were also predictive. 2023 Lavoisier. All rights reserved. -
Classification of adaptive back propagation neural network along with fuzzy logic in chronic kidney disease
A steady deterioration in kidney function over months or years is known as chronic kidney disease (CKD). Through a range of techniques, such as pharmacological intervention in moderate cases and hemodialysis and renal transport in severe cases, early identification of CKD is crucial and has a substantial influence on reducing the patient's health development. The outcomes show the patient's kidneys' present state. It is suggested to develop a system for detecting chronic renal disease using machine learning. Finding the best feature sets typically involves using metaheuristic algorithms since feature selection is an NP-hard issue with amorphous polynomials. Semi-crystalline tabu search (TS) is frequently used for both local and global searches. In this study, we employ a brand-new hybrid TS with stochastic diffusion search (SDX)-based feature selection. The adaptive backpropagation neural network (ABPNN-ANFIS) is then classified using fuzzy logic. Fuzzy logic may be used to combine the ABPNN findings. Consequently, these techniques can aid experts in determining the stage of chronic renal disease. The Adaptive Neuron Clearing Inference System (ABPNN-ANFIS) was utilised to develop adaptive inverse neural networks using the MATLAB programme. The outcomes demonstrate that the suggested ABPNN-ANFIS is 98 % accurate in terms of efficiency. 2024 -
-
Classification of Alzheimer's Disease Stages Using Machine Learning Techniques
Alzheimer s disease (AD) is a type of mental disorder which deteriorates the normal functioning of human brain by reducing the memory capacity of an individual. Age is the most common factor for AD and this disease cannot be reversed or stopped. Doctors can only treat the symptoms of AD which include personality changes and brain structural changes. Analyzing neuro-degenerative disorders, neuroimaging plays an important role in diagnosing subjects with AD and other stages of AD. The proposed research identified this gap and using MRI and PET newlineimages for recognizing AD in its early occurrences by the professionals. This helps in tailoring an appropriate treatment procedure for treating AD. As per literature survey, many researchers have worked with convolutional methods like inbuilt skull stripping with two or more conversions and classified with different CNN architectures. The proposed research experimented advanced skull stripping method and classified using deep learning architectures. This research emphasizes the implementation of ResNet50 architecture with T1 weighted MRI and Amyloid PET images for detecting the abnormalities in the brain patterns based on the image attributes. For the proposed experiment, a total of 5000 T1 weighted MRI data and 3000 newlineAmyloid PET data were used. The collected images were pre-processed with noise removal newlinetechniques and skull stripping method. The ResNet50 is used to classify AD from the data newlineobtained from the ADNI dataset. Pre-processed images /data were fed to the tuned for three class classification on ADNI image data at 200 Epochs shows the accuracy of 97.3% for T1 weighted MRI data and 98% for Amyloid PET data. The experimental results of the proposed model prove that it classifies the images according to various stages with better accuracy than the other existing models by achieving excellent results. -
Classification of Breast Invasive Ductal Carcinomas Using Histopathological Images Based on Deep Learning Techniques
Women suffer from cancer, which is the main reason for death for females around the world. With the use of artificial intelligence, it is possible to predict and detect all types of cancers in the near future. It is not just women who can heal, and most breast cancers are caused by the most vulnerable type of breast. Eighty percent of all diagnoses of carcinoma are invasive ductal carcinomas (IDCs). In this paper, deep learning techniques are extended to support visible semantic evaluation of tumor areas, using convolutional neural networks (CNNs).A CNN is skilled ended a large number of photo covers (tissue areas) after Whole Slide Images (WSI) to study ranked part-based total image. About 600 normal image patches and 200 breast invasive ductal carcinomas are selected for the experiment. It was intended to amount classifier correctness in the detection of IDC tissue areas in Whole Slide Images. We achieved excellent measurable outcomes for an automated finding of IDC areas with our technique. The results are evaluated based on performance measures and compared with a different number of neurons, and the results are highlighted. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Classification of countries based on development indices by using K-means and grey relational analysis
Clustering countries based on their development profile is important, as it helps in the efficient allocation and use of resources for institutions like the World Bank, IMF and many others. However, measuring the status of development in each country is challenging, as development encompasses several facets such as economic, social, environmental and institutional aspects. These dimensions should be captured and aggregated appropriately before attempting to classify countries based on development. In this context, this paper attempts to measure various dimensions of development through four indices namely, Economic Index (EI), Social Index (SI), Sustainability Index (SUI) and Institutional Index (II) for the period between 1996 through 2015 for 102 countries. And then we categorize the countries based on these development indices using the grey relational analysis and K-means clustering method. Our study classifies countries into four clusters with twelve countries in the first cluster, fifty in second, twenty-seven and thirteen countries in third and fourth clusters respectively. Having taken each of the dimensions of development independently, our results show that no cluster has performed poorly in all four aspects. 2021, The Author(s), under exclusive licence to Springer Nature B.V. -
Classification of Disaster Tweets using Machine Learning and Deep Learning Techniques
Social networks provide a plethora of information for gathering extra data on people's behavior, trends, opinions, and feelings during human-affecting occurrences, such as natural catastrophes. Twitter is an inevitable communication medium during calamities. People mainly depend on Twitter to announce real-time emergencies. However, it is rarely straightforward if someone is declaring a tragedy. Sentiment analysis of disaster tweets aid in situational awareness and realizing the disaster dynamics. In our paper, we perform a sentimental analysis of disaster tweets using techniques based on machine learning and deep learning. The tweets are pre-processed before being converted into a structured form using Natural Language Processing (NLP) methods. Supervised learning techniques such as the Support Vector Machine and the Naive Bayes Classifier algorithm are used to develop the Classifier, which categorizes tweets into distinct catastrophes and selects the most appropriate algorithm. The chosen algorithm is further enriched with an emoticon detection algorithm for explicit elucidation. Our research would help disaster relief organizations and news agencies to conclude about the state of affairs and do the needful. 2022 IEEE. -
Classification of Diseased Leaves in Plants Using Convolutional Neural Networks
The article focuses on the classification of diseased leaves using a machine learning algorithm. The main focus in agriculture is controlling pests and weeds, for which farmers spray chemical pesticides to get a good yield. The issue here is over-usage and under-usage of pesticides, which might harm the end consumer. To achieve the goal of reducing pesticide use and detecting pests in the crop early, the machine learning algorithm is deployed on the leaf image. The image data of the leaf of the cauliflower plant is collected for 40days. The data was collected from the day the plant was seeded in a pot until the day it was ready to be planted in the soil. From this data, the pest attack on the plants is tracked without the application of pesticides. To achieve this, the CNN algorithm is used on the collected image data. The outcome of the study would be to classify the diseased leaves based on the pest attack and know the right time to spray the pesticides to reduce the damage to the plant. This also reduces the use of pesticides and costs to the farmer. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Classification of epileptic seizures using wavelet packet log energy and norm entropies with recurrent Elman neural network classifier
Electroencephalogram shortly termed as EEG is considered as the fundamental segment for the assessment of the neural activities in the brain. In cognitive neuroscience domain, EEG-based assessment method is found to be superior due to its non-invasive ability to detect deep brain structure while exhibiting superior spatial resolutions. Especially for studying the neurodynamic behavior of epileptic seizures, EEG recordings reflect the neuronal activity of the brain and thus provide required clinical diagnostic information for the neurologist. This specific proposed study makes use of wavelet packet based log and norm entropies with a recurrent Elman neural network (REN) for the automated detection of epileptic seizures. Three conditions, normal, pre-ictal and epileptic EEG recordings were considered for the proposed study. An adaptive Weiner filter was initially applied to remove the power line noise of 50Hz from raw EEG recordings. Raw EEGs were segmented into 1s patterns to ensure stationarity of the signal. Then wavelet packet using Haar wavelet with a five level decomposition was introduced and two entropies, log and norm were estimated and were applied to REN classifier to perform binary classification. The non-linear Wilcoxon statistical test was applied to observe the variation in the features under these conditions. The effect of log energy entropy (without wavelets) was also studied. It was found from the simulation results that the wavelet packet log entropy with REN classifier yielded a classification accuracy of 99.70% for normal-pre-ictal, 99.70% for normal-epileptic and 99.85% for pre-ictal-epileptic. 2016, Springer Science+Business Media Dordrecht. -
Classification of extragalactic point sources and flux variability characteristics of blazars
Classification of different types of astronomical objects in large surveys usually done through spectroscopy requires enormous amounts of time. Hence, many attempts have been made using broad band photometric magnitudes and spectroscopic observations to classify the sources, particularly extragalactic sources such as active galactic nuclei (AGNs), starburst galaxies and normal galaxies. However, a method which does not involve spectroscopic data would be ideal. -
Classification of extragalactic point sources and flux variability characteristics of blazars
Classification of different types of astronomical objects in large surveys usually done through spectroscopy requires enormous amounts of time. Hence, many attempts have been made using broad band photometric magnitudes and spectroscopic observations to classify the sources, particularly extragalactic sources such as active galactic nuclei (AGNs), starburst galaxies and newlinenormal galaxies. However, a method which does not involve spectroscopic data would be ideal. With this in view, in this work we have made an effort to classify a sample of 37,492 point sources into Quasi-Stellar Objects (QSOs), galaxies and stars using template fitting technique and multiwavelength photometric magnitudes from the Sloan Digital Sky Survey (SDSS) and newlinethe Galaxy Evolution Explorer (GALEX) with coverage from the optical (z: 8931 to the far ultraviolet (FUV: 1516 . Templates for QSOs, galaxies and stars were used to fit the data of the objects to the seven photometric bands of SDSS and GALEX. The results were compared with SDSS spectroscopic classification. Two UV bands (NUV and FUV) were included to remove the possible degeneracies in the classification based only on optical bands or in color-color method. UV bands play a crucial role in the classification and characterization of astronomical objects that emit over a wide range of wavelengths, especially for those that are bright at UV. Classification using template fitting method is consistent with spectroscopic methods, provided UV information of the objects is available. UV bands are particularly important for separating quasars and stars, as well as spiral and starburst galaxies. We have achieved the efficiency of 89% for QSOs, 63% for galaxies and 84% for stars. Objects for which spectroscopic data is not available can also be classified using this method which does not require spectroscopic information. -
Classification of fibroid using novel fully connected CNN with back propagation classifier (NFCCNNBP)
In this phase, we utilize features extracted from a prior stage to classify uterine fibroids. We employ a predefined dataset with feature values as our training set for a novel classifier called the "Novel Fully Connected CNN with Back Propagation Classifier."This classifier learns from the training set. We then put this method to the test with new images not included in the training dataset. Its primary objective is to assess the extent of infection across the entire uterine surface. Through the adoption of a Convolutional Neural Network (CNN) combined with Back Propagation (BP), we have achieved an impressive accuracy rate of 98.3% for predictions. When we compare this accuracy to existing classifiers like Fuzzy Logic, Naive Bayes, and SVM, our proposed model, NFCCNNBP, outperforms them significantly. 2024 Author(s). -
Classification of financial news articles using machine learning algorithms
The opinion helps in determining the direction of the stock market. Information hidden in news articles is an information treasure which needs to be extracted. The present study is conducted to explore the application of text mining in binning the financial articles according to the opinion expressed inside them. It is discovered that using the tri-n-gram feature extraction process in conjugation with Support Vector machines increases the reliability and precision of the binning process. The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2021. -
Classification of HHO-based Machine Learning Techniques for Clone Attack Detection in WSN
Thanks to recent technological advancements, low-cost sensors with dispensation and communication capabilities are now feasible. As an example, a Wireless Sensor Network (WSN) is a network in which the nodes are mobile computers that exchange data with one another over wireless connections rather than relying on a central server. These inexpensive sensor nodes are particularly vulnerable to a clone node or replication assault because of their limited processing power, memory, battery life, and absence of tamper-resistant hardware. Once an attacker compromises a sensor node, they can create many copies of it elsewhere in the network that share the same ID. This would give the attacker complete internal control of the network, allowing them to mimic the genuine nodes' behavior. This is why scientists are so intent on developing better clone assault detection procedures. This research proposes a machine learning based clone node detection (ML-CND) technique to identify clone nodes in wireless networks. The goal is to identify clones effectively enough to prevent cloning attacks from happening in the first place. Use a low-cost identity verification process to identify clones in specific locations as well as around the globe. Using the Optimized Extreme Learning Machine (OELM), with kernels of ELM ideally determined through the Horse Herd Metaheuristic Optimization Algorithm (HHO), this technique safeguards the network from node identity replicas. Using the node identity replicas, the most reliable transmission path may be selected. The procedure is meant to be used to retrieve data from a network node. The simulation result demonstrates the performance analysis of several factors, including sensitivity, specificity, recall, and detection. 2023, Modern Education and Computer Science Press. All rights reserved. -
Classification of Hypothyroid Disorder using Optimized SVM Method
Hypothyroidism is an endocrine disorder where the thyroid organ doesn't provide the enough amount of thyroid hormones. It is one of the common diseases found in women. Detection of hypothyroidism needs suitable diagnostic tests to encourage prompt analysis and medication. Accurate and early detection of a disease is more important and compulsory in healthcare domain to facilitate correct and prompt diagnosis and timely treatment. The information generated in healthcare domain is on large scale, crucial and complex with multiple parameters. To interpret and understand such a huge data and retrieve the accurate and relevant information from it is a tedious as well as challenging task. However, there is a need and importance to facilitate the patients with better medical solutions. This will help to reduce the cost, time and give more relief to users by applying advanced and upgraded knowledge. It will also assist to prevent the further complications. The proposed study gains the knowledge from the hypothyroid dataset to predict the level of disease. To identify the level of hypothyroid disorder, we used four classification machine learning techniques, namely KNN (K-Nearest Neighbour), SVM (Support Vector Machines), LR (Logistic Regression) and NN (Artificial Neural Network). The Experimental results compared the classification accuracy of four methods. Logistic Regression method achieved 96.08% accuracy among other three classifiers. But, SVM is found the best classifier after standardizing the data and parameter tuning with accuracy of 99.08%. 2019 IEEE. -
Classification of myocardial ischemia in delayed contrast enhancement using machine learning
This chapter addresses the classification of myocardial ischemia in delayed contrast enhancement using machine-learning techniques for magnetic resonance imaging which solves the social issue of a sudden cardiac death. To automate the classification of myocardial ischemia, the computer-aided design has a crucial path on the mixture ensemble of machine learning. The mixture ensemble of machine learning can partition a high-dimensional image in a simultaneous and competitive way. The detection and the segmentation processes are carried out through Fuzzy C-Means multispectral and single-channel algorithms along with a morphological filtering technique for feature extraction. Furthermore, the feed forward neural network (FFNN) technique is trained through the Levenberg-Marquardt Back Propagation algorithm for the classification of myocardial ischemia in delayed contrast enhancement. The proposed classification model performs well for the classification of myocardial ischemia. The rigorous process of the proposed result reveals that the FFNN classifier produces 99.9% accuracy on the classification of myocardial ischemia and also shows that the given classifier is considered one of the best methods in classifying medical myocardial ischemia. 2019 Elsevier Inc. All rights reserved. -
Classification of Psychological Disorders by Feature Ranking and Fusion using Gradient Boosting Classification of Psychological Disorders
Negative emotional regulation is a defining element of psychological disorders. Our goal was to create a machine-learning model to classify psychological disorders based on negative emotions. EEG brainwave dataset displaying positive, negative, and neutral emotions. However, negative emotions are responsible for psychological health. In this paper, research focused solely on negative emotional state characteristics for which the divide-and-conquer approach has been applied to the feature extraction process. Features are grouped into four equal subsets and feature selection has been done for each subset by feature ranking approach based on their feature importance determined by the Random Forest-Recursive Feature Elimination with Cross-validation (RF-RFECV) method. After feature ranking, the fusion of the feature subset is employed to obtain a new potential dataset. 10-fold cross-validation is performed with a grid search created using a set of predetermined model parameters that are important to achieving the greatest possible accuracy. Experimental results demonstrated that the proposed model has achieved 97.71% accuracy in predicting psychological disorders 2023, International Journal of Advanced Computer Science and Applications.All Rights Reserved. -
Classification of Silicon (Si) Wafer Material Defects in Semiconductor Choosers using a Deep Learning ShuffleNet-v2-CNN Model
The silicon wafer is one of the raw materials used to make semiconductor chipsets. Semiconductor failure or dysfunction could be the result of defects in the layers of this material. As a result, it is essential to work toward the development of a system that is both quick and precise in identifying and classifying wafer defects. Wafer map analysis is necessary for the quality control and analysis of the semiconductor manufacturing process. There are some failure patterns that can be displayed by wafer maps. These patterns can provide essential details that can assist engineers in determining the reason of wafer failures. In this research, a deep-learning-based silicon wafer defect identification and classification model is proposed. The main objective of this research is to identify and classify the silicon wafer defects using the wafer map images. This proposed model identifies and classifies the defects based on the wafer map images from the WM-811K dataset. The proposed model is composed of a pretrained deep transfer learning model called ShuffleNet-v2 with convolutional neural network (CNN) architecture. This ShuffleNet-v2-CNN performs the defects identification and classification process following the workflow of data preprocessing, data augmentation, feature extraction, and classification. For performance evaluation, the proposed ShuffleNet-v2-CNN is evaluated with performance metrics like accuracy, recall, precision, and f1-score. The proposed model has obtained an overall accuracy of 96.93%, 95.40% precision, 96.26% recall, and 95.75% F1-score in classifying the silicon wafer defects based on the wafer map images. 2022 Rajesh Doss et al. -
Classification of Skin Diseases Using Convolutional Neural Networks (VGG) with Histogram Equalization Preprocessing
Skin diseases are a major global health concern for which prompt and precise diagnosis is necessary for effective treatment. Convolutional neural networks (CNN), one of the deep learning techniques, have shown potential in automating the diagnostic procedure. The goal of this research is to enhance the effectiveness of skin disease categorization by fusing the capabilities of CNNs - particularly the VGG architecture - with the histogram equalization preprocessing method. In image processing, histogram equalization is a commonly used approach to enhance the contrast and general quality of medical photographs, which include photos of skin conditions. In order to improve the characteristics and details of dermatological pictures for this study, we employed histogram equalization as a preprocessing step. This allowed CNN to extract pertinent features more quickly. 2024 IEEE.