Search

Sentiment analysis is a process in Natural Language Processing that involves detecting and classifying emotions in texts. The emotion is focused on a specific thing, an object, an incident, or an individual. Although some tasks are concerned with detecting the existence of emotion in text, others are concerned with finding the polarities of the text, which is classified as positive, negative, or neutral. The task of determining whether a comment contains inappropriate text that affects either individual or group is called offensive language identification. The existing research has concentrated more on sentiment analysis and offensive language identification in a monolingual data set than code-mixed data. Code-mixed data is framed by combining words and phrases from two or more distinct languages in a single text. It is quite challenging to identify emotion or offensive terms in the comments since noise exists in code-mixed data. The majority of advancements in hostile language detection and sentiment analysis are made on monolingual data for languages with high resource requirements. The proposed system attempts to perform both sentiment analysis and offensive language identification for low resource code-mixed data in Tamil and English using machine learning, deep learning and pre-trained models like BERT, RoBERTa and adapter-BERT. The dataset utilized for this research work is taken from a shared task on Multi task learning DravidianLangTech@ACL2022. Another challenge addressed by this work is the extraction of semantically meaningful information from code-mixed data using word embedding. The result represents an adapter-BERT model gives a better accuracy of 65% for sentiment analysis and 79% for offensive language identification when compared with other trained models., (© 2022. This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply.)

Breast cancer is a commonly occurring disease in women all over the world. Mammogram is an efficient technique used for screening and identification of abnormalities over the breast region. Earlier identification of breast cancer enhances the prognosis of patients and is mainly based on the experience of the radiologist in interpretation of mammogram with quality of image. The advent of Deep Learning (DL) and Computer Vision techniques is widely used to perform breast cancer diagnosis. This paper presents a new Optimal Multi-Level Thresholding-based Segmentation with DL enabled Capsule Network (OMLTS-DLCN) breast cancer diagnosis model utilizing digital mammograms. The OMLTS-DLCN model involves an Adaptive Fuzzy based median filtering (AFF) technique as a pre-processing step to eradicate the noise that exists in the mammogram images. Besides, Optimal Kapur's based Multilevel Thresholding with Shell Game Optimization (SGO) algorithm (OKMT-SGO) is applied for breast cancer segmentation. In addition, the proposed model involves a CapsNet based feature extractor and Back-Propagation Neural Network (BPNN) classification model is employed to detect the existence of breast cancer. The diagnostic outcomes of the presented OMLTS-DLCN technique is examined by means of benchmark Mini-MIAS dataset and DDSM dataset. The experimental values obtained highlights the superior performance of the OMLTS-DLCN model with a higher accuracy of 98.50 and 97.55% on the Mini-MIAS dataset and DDSM dataset, respectively., (© 2021. International Association of Scientists in the Interdisciplinary Areas.)

Machine learning (ML) is currently a crucial tool in the field of cyber security. Through the identification of patterns, the mapping of cybercrime in real time, and the execution of in-depth penetration tests, ML is able to counter cyber threats and strengthen security infrastructure. Security in any organization depends on monitoring and analyzing user actions and behaviors. Due to the fact that it frequently avoids security precautions and does not trigger any alerts or flags, it is much more challenging to detect than traditional malicious network activity. ML is an important and rapidly developing anomaly detection field in order to protect user security and privacy, a wide range of applications, including various social media platforms, have incorporated cutting-edge techniques to detect anomalies. A social network is a platform where various social groups can interact, express themselves, and share pertinent content. By spreading propaganda, unwelcome messages, false information, fake news, and rumours, as well as by posting harmful links, this social network also encourages deviant behavior. In this research, we introduce Deep Belief Network (DBN) with Triple DES, a hybrid approach to anomaly detection in unbalanced classification. The results show that the DBN-TDES model can typically detect anomalous user behaviors that other models in anomaly detection cannot. [ABSTRACT FROM AUTHOR]

Human activity recognition (HAR) is one of the key applications of health monitoring that requires continuous use of wearable devices to track daily activities. The most efficient supervised machine learning (ML)-based approaches for predicting human activity are based on a continuous stream of sensor data. Sensor data analysis for human activity recognition using conventional algorithms and deep learning (DL) models shows promising results, but evaluating their ambiguity in decision-making is still challenging. In order to solve these issues, the paper proposes a novel Wasserstein gradient flow legonet WGF-LN-based human activity recognition system. At first, the input data is pre-processed. From the pre-processed data, the features are extracted using Haar Wavelet mother- Symlet wavelet coefficient scattering feature extraction (HS-WSFE). After that, the interest features are selected from the extracted features using (Binomial Distribution integrated-Golden Eagle Optimization) BD-GEO. The important features are then post-processed using the scatter plot matrix method. Obtained post-processing features are finally given into the WGF-LN for classifying human activities. From these experiments, the results can be obtained and showed the efficacy of the proposed model. [ABSTRACT FROM AUTHOR]

Fingerprint recognition has emerged as one of the most reliable biometric authentication methods, owing to its uniqueness and permanence. However, the security and confidentiality of the user's data are key considerations in modern biometric systems. In this study, we describe an intelligent computational technique for automatically validating fingerprints for identification and verification purposes. The feature vector is created by fusing Gabor filtering features with deep learning techniques like the faster region-based convolutional neural network (Faster R-CNN). This study uses linear and decision-based median filtering (DBMF) techniques to minimize visual impulse noise. Faster-R-CNN with DBMF was applied to the feature vectors to reduce overfitting problems while improving classification precision and reliability. For fingerprint matching, the Euclidean distance between the associated Harris-SURF feature vectors of two feature points is used to measure feature-matching similarity between two fingerprint images. Furthermore, for fine-tuned matching an iterative technique known as RANSAC (Random Sample Consensus) is used. The experimental results collected from the public-domain fingerprint databases FVC-2002 DB1 and FVC-2000 DB1 show that the proposed design is viable and performs well with an accuracy of 99.43%, MSE value of 43.321%, and an execution time of 3.102 ms which was more exact than existing models. • Reduces image impulse noise using decision-based median and linear filtering. • Feature vectors were processed with Faster-R-CNN and DBMF to alleviate overfitting. • Optimizes the matching procedure and automatically eliminates incorrect matches. • Experimentation with publicly available datasets of FVC2002 DB1 and FVC2000 DB1. [ABSTRACT FROM AUTHOR]

Automated Speech Emotion Recognition (SER) becomes more popular and has increased applicability. SER concentrates on the automatic identification of the emotional state of a human being using speech signals. It mainly depends upon the in-depth analysis of the speech signal, extracts features containing emotional details from the speech signal, and utilises pattern recognition techniques for emotional state identification. The major problem in automatic SER is to extract discriminate, powerful, and emotional salient features from the acoustical content of speech signals. The proposed model aims to detect and classify three emotional states of speech such as happy, neutral, and sad. The presented model makes use of Convolution neural network – Gated Recurrent unit (CNN-GRU) based feature extraction technique which derives a set of feature vectors. A comprehensive simulation takes place using the Berlin German Database and SJTU Chinese Database which comprises numerous audio files under a collection of different emotion labels. [ABSTRACT FROM AUTHOR]

With the advancement of communication and computing technologies, multimedia technologies involving video and image applications have become an important part of the information society and have become inextricably linked to people's daily productivity and lives. Simultaneously, there is a growing interest in super-resolution (SR) video reconstruction techniques. At the moment, the design of digital twins in video computing and video reconstruction is based on a number of difficult issues. Although there are several SR reconstruction techniques available in the literature, most of the works have not considered the spatiotemporal relationship between the video frames. With this motivation in mind, this paper presents VDCNN-SS, a novel very deep convolutional neural networks (VDCNN) with spatiotemporal similarity (SS) model for video reconstruction in digital twins. The VDCNN-SS technique proposed here maps the relationship between interconnected low resolution (LR) and high resolution (HR) image blocks. It also considers the spatiotemporal non-local complementary and repetitive data among nearby low-resolution video frames. Furthermore, the VDCNN technique is used to learn the LR-HR correlation mapping learning process. A series of simulations were run to examine the improved performance of the VDCNN-SS model, and the experimental results demonstrated the superiority of the VDCNN-SS technique over recent techniques. [ABSTRACT FROM AUTHOR]

One of the deadliest diseases in the world is brain cancer. Children and adults are also susceptible to this malignancy. It also has the poorest rate of survival and comes in a variety of shapes, textures, and sizes, depending on where it is found. Bad things will happen if the tumour brain is misclassified. As a reason, early detection of the right kind and grade of tumour is critical in determining the best treatment strategy. Brain tumours may be identified by looking at magnetic resonance imaging (MRI) pictures of the patient's brain. The manual method becomes time-consuming and may lead to human mistakes due to the huge quantities of data and the different kinds of brain tumours. As a result, a computer-assisted diagnostic (CAD) system is needed. Image categorization methods have advanced significantly in recent years, particularly deep learning networks, which have achieved success in this field. In this case, we used a multilayer stacked probabilistic belief network to accurately classify brain tumors. Here the MRI brain images are Pre-processed using the Hybrid Butter worth Anisotropic filter and contrast Blow up Histogram Equalization. Followed by pre-processing, the denoised image can be segmented by using the bounding box U-NET segmentation methodology. Then after segmenting the target, the specialized features regarding the tumor can be extracted using the In-depth atom embedding method. Then they obtained can reduce feature dimensionality by using the Backward feature eliminating green wing optimization. The extracted features can be given as input for the classification process. A Multilayer stacked probabilistic belief network is then used to classify the tumour as malignant or benign. The suggested system's efficacy was tested on the BraTS dataset, which yielded a high level of accuracy. Subjective comparison study is also performed out among the suggested technique and certain state-of-the-art methods, according to the work presented. Experiments show that the proposed system outperforms current methods in terms of assisting radiologists in identifying the size, shape, and location of tumors in the human brain. [ABSTRACT FROM AUTHOR]

In recent years, web crawling has gained a significant attention due to the drastic advancements in the World Wide Web. Web Search Engines have the issue of retrieving massive quantity of web documents. One among the web crawlers is the focused crawler, that intends to selectively gather web pages from the Internet. But the efficiency of the focused crawling can easily be affected by the environment of web pages. In this view, this paper presents an Automated Word Embedding with Parameter Tuned Deep Learning (AWE-PTDL) model for focused web crawling. The proposed model involves different processes namely pre-processing, Incremental Skip-gram Model with Negative Sampling (ISGNS) based word embedding, bidirectional long short-term memory-based classification and bird swarm optimization based hyperparameter tuning. The SGNS training desires to go over the complete training data to pre-compute the noise distribution before performing Stochastic Gradient Descent (SGD) and the ISGNS technique is derived for the word embedding process. Besides, the cosine similarity is computed from the word embedding matrix to generate a feature vector which is fed as input into the Bidirectional Long Short-Term Memory (BiLSTM) for the prediction of website relevance. Finally, the Birds Swarm Optimization-Bidirectional Long Short-Term Memory (BSO-BiLSTM) based classification model is used to classify the webpages and the BSO algorithm is employed to determine the hyperparameters of the BiLSTM model so that the overall crawling performance can be considerably enhanced. For validating the enhanced outcome of the presented model, a comprehensive set of simulations are carried out and the results are examined in terms of different measures. The Automated Word Embedding with Parameter Tuned Deep Learning (AWE-PTDL) technique has attained a higher harvest rate of 85% when compared with the other techniques. The experimental results highlight the enhanced web crawling performance of the proposed model over the recent state of art web crawlers. [ABSTRACT FROM AUTHOR]

In recent years, intelligent automation in the healthcare sector becomes more familiar due to the integration of artificial intelligence (AI) techniques. Intelligent healthcare systems assist in making better decisions, which further enable the patient to provide improved medical services. At the same time, skin lesion is a deadly disease that affects people of all age groups. Skin lesion segmentation and classification play a vital part in the earlier and precise skin cancer diagnosis by intelligent systems. However, the automated diagnosis of skin lesions in dermoscopic images is challenging because of the problems such as artifacts (hair, gel bubble, ruler marker), blurry boundary, poor contrast, and variable sizes and shapes of the lesion images. This study develops intelligent multilevel thresholding with deep learning (IMLT-DL) based skin lesion segmentation and classification model using dermoscopic images to address these problems. Primarily, the presented IMLT-DL model incorporates the Top hat filtering and inpainting technique for the pre-processing of the dermoscopic images. In addition, the Mayfly Optimization (MFO) with multilevel Kapur’s thresholding-based segmentation process is involved in determining the infected regions. Besides, an Inception v3 based feature extractor is applied to derive a valuable set of feature vectors. Finally, the classification process is carried out using a gradient boosting tree (GBT) model. The presented model’s performance takes place against the International Skin Imaging Collaboration (ISIC) dataset, and the experimental outcomes are inspected in different evaluation measures. The resultant experimental values ensure that the proposed IMLT-DL model outperforms the existing methods by achieving higher accuracy of 0.992. [ABSTRACT FROM AUTHOR]

Rapid industrialization has led to the generation of a considerable amount of waste, both solid and liquid, in industrial fields like food processing, sugar, pulp, sago or starch, dairies, paper, fruit processing, poultry, distilleries, slaughterhouses, tanneries, and so forth. Despite the requirement for pollution control measures, the waste is discharged into water bodies or generally dumped on land without appropriate management, and thus becomes a significant source of environmental pollution and health hazards. The most essential step of waste management is the segregation of waste into the various elements, and normally this process is done automatically by hand-picking. A smart waste material classification technique is required to simplify the procedures. Therefore, the study presents a new Metaheuristics with Deep Transfer Learning Enabled Detection and Classification Methods for Industrial Waste Management (MDTLDC-IWM) method. The presented MDTLDC-IWM model facilitates the use of DL models for the identification and classification of waste materials in the IWM system. To accomplish this, the presented MDTLDC-IWM model follows two key phases, namely waste object recognition and waste object classification. At the initial stage, the YOLO-v5 object detector with the Harris Hawks Optimization (HHO) algorithm is used. Next, in the second stage, the stacked sparse auto encoder (SSAE) model is applied for the waste object classification method. The SSAE model is effectively optimized using the Aquila Optimization Algorithm (AOA), which helps to accomplish maximum classification of waste objects. The MDTLDC-IWM model has achieved a precision of 96.84 percent and an F score of 96.71 percent. A benchmark dataset is used to test the experimental validity of the presented MDTLDC-IWM model. Extensive comparative analysis reported the enhanced performance of the MDTLDC-IWM model over recent state-of-the-art approaches. [Display omitted] • This study introduces a novel Metaheuristics with Deep Transfer Learning Enabled Detection and Classification Methods for Industrial Waste Management (MDTLDC-IWM) approach. • The described MDTLDC-IWM model supports the employment of DL models in the IWM system for the identification and classification of waste products. Since the YOLO-v5 object detector with the Harris Hawks Optimization (HHO) method is employed. • The stacked sparse auto encoder (SSAE) model is utilised for the classification of waste objects. [ABSTRACT FROM AUTHOR]

Recent advancements in renewable energy provide a clean, alternative source to fossil fuels. Smart grids (SGs) work by collecting data about customer requests, comparing them to current supply data, computing electricity costs, and so on. Because the processes are time-dependent, a dynamic estimate of SG stability is a critical system need. Analyzing fluctuations and disturbances in a dynamic manner is critical for the SGs to work properly. Recent advancements in data science, machine learning (ML), and deep learning (DL) models have facilitated the creation of useful stability prediction models for the SGs environment. In this regard, this study provides an Artificial Hummingbird (AHB) algorithm-based feature selection model for the SG environment with optimal DL enabled stability prediction (AHBFS-ODLSP). The AHBFS-ODLSP model is primarily concerned with the design of an AHB-based feature selection technique. Furthermore, a prediction system based on Multiheaded-Self Attention Long Short-Term Memory (MHSA-LSTM) is being created right now to predict the stability level. Then, using the symbiotic organism search (SOS) optimization technique, the MHSA-LSTM model hyperparameters were adjusted. The AHBFS and SOS algorithm designs have a big impact on how well the MHSA-LSTM model predicts stability. AHBFS-ODLSP model modifications are demonstrated through a number of simulations, and the outcomes are assessed in a variety of ways. The AHBFS-ODLSP method has achieved its maximum performance with a F score of 99.02 %. The AHBFS-ODLSP model outperformed other prediction models, according to a thorough comparative analysis. • Artificial Humming Bird (AHB) algorithm based feature selection with optimal DL enabled stability prediction. • AHBFS-ODLSP model encompasses the design of AHB based feature selection technique. • MHSA-LSTM based prediction method is developed to predict the level of stability. • SOS optimization technique was employed to fine tune the hyperparameter related to the MHSA-LSTM model. [ABSTRACT FROM AUTHOR]

• Fuzzy logic-based energy management based new traffic flow model was applied in the cyber physical system. • This research focuses on the construction of a fuzzy logic-based energy management and TFP model (FLEM-TFP) for CPS in ITS with this rationale. • An adaptive neuro fuzzy inference system (ANFIS) model is used to calculate the engine torque necessary depending on various measurements. • The FLEM-TFP technique combines two distinct processes: optimal ANFIS-based energy management and a TFP model based on SFO. • The SFO model is created by training fuzzy wavelet neural network (FWNN) with the SFO algorithm to predict traffic flow in an ITS environment. Cyber-Physical Systems (CPS) is a multidisciplinary effort that links cyber and physical vectors. Cyber systems may be incorporated into any physical system, such as intelligent transportation systems (ITS), industries, and cities, to improve its intelligence, energy efficiency, and comfort. The design of self-driving cars has a large influence on vehicle travel demand and energy consumption. ITS also heavily relies on the Traffic Flow Prediction (TFP) system. With this rationale, this work focuses on the development of a Fuzzy Logic-based Energy Management and TFP model (FLEM-TFP) for CPS in ITS. The FLEM-TFP approach proposed here involves two primary processes: energy management and TFP. An adaptive neuro-fuzzy inference system (ANFIS) model is also utilized to compute the engine torque required based on various readings. In addition, a sailfish optimizer (SFO)-based fuzzy wavelet neural network (FWNN) approach is used in the ITS to estimate traffic flow. According to the findings of the experiments, the FLEM-TFP approach outperformed previous state-of-the-art procedures. [Display omitted] [ABSTRACT FROM AUTHOR]

• Decision Tree, SVM, and Naive Bayes classification devices were used by Deepti and Dilip to diagnose diabetes. • The goal was to classify large amounts of the separator. PIMA Indian database was used. • Naive Bayes, scored 76.30 per cent, received the best accuracy. Six separators were used. • The LIBSVM wallet F select script selects four characteristics and selects 9 and 20 characteristics. • The sugar prediction can be more accurate with 72 percent precision. Diabetes mellitus is a disease commonly called Diabetes. Diabetes is among the most frequent diseases globally. This disease affects internationally with different ailments and complications in majority of peoples. Diabetes is a chronic disease with the ability to create a global medical care crisis. In compliance with International Diabetes Federation 382 million people are reportedly living with diabetes throughout the entire world. The diabetes disorder can't be cured but it can control, diagnosis and prediction of diabetes is vital to restrain the death rate as a result of its seriousness globally. Many In recent years, machine learning (ML) algorithms have been used in the prediction of diabetes. A clever predictive model utilizing deep modelling is commonly advised with the aid of conditional data collection to forecast the severity and particular risk factor of diabetics. In this article, to resolve this problem we employed the Interpretable Filter based Convolutional Neural Network (IF-CNN) prediction model and Pet Dog-Smell Sensing (PD-SS) algorithm that can automatically predict the diabetes from PIMA Indian diabetes datasets. This may enhance the general strategy of disease prediction in patients database that may solve the issues faced by traditional algorithms employing the Deep Neural Network (DNN) methods. The automated extraction, selection and classification of attributes, disease forecast is the hard task with aggressive performance for your PIMA information which may be implemented with the projected Deep Learning version economically. This may enhance the general plan of disease prediction from patients database that may solve the issues faced by traditional algorithms employing the Interpretable Filter established Deep Learning version prediction model to diagnose and predict the diabetes disease in multi-level databases. The intention of this research is to create a method that could carry out early diabetes predictions for a more reliable patient by including findings of SVM and CNN-LSTM(Long Short-Term Memory) machine learning methods also IF-CNN achieved 96.26% accuracy. [ABSTRACT FROM AUTHOR]

Recently, computer vision-based face image analysis has sparked considerable interest in a variety of applications such as surveillance, security, biometrics and so on. The goal of the facial analysis was to derive facial soft biometrics such as identification, gender, age, ethnicity, expression and so on. Among these, ethnicity recognition remains a hot study topic, a major aspect of society with profound linkages to a variety of environmental and social concerns. The introduction of machine learning (ML) and deep learning (ML) technologies has proven advantageous for effective ethnicity recognition and classification. In this regard, the IDL-ERCFI technique, which is based on intelligent DL, is designed in this paper. The IDL-ERCFI technique's purpose is to distinguish and classify ethnicity based on facial photos. The IDL-ERCFI technique uses face landmarks to align photos before sending them to the network. Furthermore, the proposed model employs an Exception network as a feature extractor. Because the retrieved features are high-dimensional, the feature reduction procedure employs the principal component analysis (PCA) technique, which is effective in overcoming the "curse of dimensionality." Furthermore, the ethnicity classification procedure is carried out using an optimal kernel extreme learning machine (KELM), with parameter tuning of the KELM model carried out using the glow worm swarm optimization (GSO) technique. A complete experimental analysis is carried out to demonstrate the superiority of the IDL-ERCFI technique over the other techniques. • The IDL-ERCFI technique's purpose is to distinguish and classify ethnicity based on facial photos. • The IDL-ERCFI technique uses face landmarks to align photos before sending them to the network. • The feature retrieval, the feature reduction procedure employs the principal component analysis (PCA) model. • The ethnicity classification procedure is carried out using KELM, with a parameter tuning process. • KELM model carried out experimental analysis with the glow worm swarm optimization (GSO) technique. [ABSTRACT FROM AUTHOR]