Cardiovascular disease prediction aids practitioners in making more accurate health decisions for their patients. Early detection can aid people in making lifestyle changes and, if necessary, ensuring effective medical care. Machine learning (ML) is a plausible option for reducing and understanding heart symptoms of disease. The chi-square statistical test is performed to select specific attributes from the Cleveland heart disease (HD) dataset. Support vector machine (SVM), Gaussian Naive Bayes, logistic regression, LightGBM, XGBoost, and random forest algorithm have been employed for developing heart disease risk prediction model and obtained the accuracy as 80.32%, 78.68%, 80.32%, 77.04%, 73.77%, and 88.5%, respectively. The data visualization has been generated to illustrate the relationship between the features. According to the findings of the experiments, the random forest algorithm achieves 88.5% accuracy during validation for 303 data instances with 13 selected features of the Cleveland HD dataset.

Social media is Internet-based by design, allowing people to share content quickly via electronic means. People can openly express their thoughts on social media sites such as Twitter, which can then be shared with other people. During the recent COVID-19 outbreak, public opinion analytics provided useful information for determining the best public health response. At the same time, the dissemination of misinformation, aided by social media and other digital platforms, has proven to be a greater threat to global public health than the virus itself, as the COVID-19 pandemic has shown. The public's feelings on social distancing can be discovered by analysing articulated messages from Twitter. The automated method of recognizing and classifying subjective information in text data is known as sentiment analysis. In this research work, we have proposed to use a combination of preprocessing approaches such as tokenization, filtering, stemming, and building N-gram models. Deep belief neural network (DBN) with pseudo labelling is used to classify the tweets. Top layers of the base classifiers are boosted in the pseudo labelling strategy, whereas lower levels of the base classifiers share weights for feature extraction. By introducing the pseudo boost mechanism, our suggested technique preserves the same time complexity as a DBN while achieving fast convergence to optimality. The pseudo labelling improves the performance of the classification. It extracts the keywords from the tweets with high precision. The results reveal that using the DBN classifier in conjunction with the bigram in the N-gram model outperformed other models by 90.3 percent. The proposed approach can also aid medical professionals and decision-makers in determining the best course of action for each location based on their views regarding the pandemic.

Due to the exponential growth of high-quality fake photos on social media and the Internet, it is critical to develop robust forgery detection tools. Traditional picture- and video-editing techniques include copying areas of the image, referred to as the copy-move approach. The standard image processing methods physically search for patterns relevant to the duplicated material, restricting the usage in enormous data categorization. On the contrary, while deep learning (DL) models have exhibited improved performance, they have significant generalization concerns because of their high reliance on training datasets and the requirement for good hyperparameter selection. With this in mind, this article provides an automated deep learning-based fusion model for detecting and localizing copy-move forgeries (DLFM-CMDFC). The proposed DLFM-CMDFC technique combines models of generative adversarial networks (GANs) and densely connected networks (DenseNets). The two outputs are combined in the DLFM-CMDFC technique to create a layer for encoding the input vectors with the initial layer of an extreme learning machine (ELM) classifier. Additionally, the ELM model’s weight and bias values are optimally adjusted using the artificial fish swarm algorithm (AFSA). The networks’ outputs are supplied into the merger unit as input. Finally, a faked image is used to identify the difference between the input and target areas. Two benchmark datasets are used to validate the proposed model’s performance. The experimental results established the proposed model’s superiority over recently developed approaches.

No abstract

Wireless sensor networks find applications everywhere in day to day activities right from attendance entry systems to healthcare monitoring systems. The evolution of the Internet of Things (IoT) as the Internet of Everything (IoET) makes the wireless sensor network omnipresent and increases the use of Radio Frequency Identification (RFID) for the proper identification of devices and sensor nodes which are mostly battery operated. As technology evolves, security threats also increase rapidly. This mandates a strong and energy-efficient green solution. This work attempted to address these issues by effectively deploying the lightweight encryption scheme called Extended Tiny Encryption Algorithm (XTEA). Though the XTEA is lightweight and famous, it is commonly known for various attacks. Our work patches the security threats in the XTEA by applying domain-specific customization, random number utilization, and undisclosed key renewal techniques. Two custom Renovated XTEA Mutual Authentication Protocol (RXMAP) encoder architectures, namely, RXMAP-1 and RXMAP-2, are proposed based on the replacement of accurate computational blocks with approximate blocks. The proposed RXMAP protocol is evaluated for its computational and storage overhead and verified against various security threats using BAN logic formal verification and informal verification. The proposed encoder architectures are simulated for functional verification, and ASIC implementation is done with a 132 nm process node. ASIC implementation results show that the proposed designs RXMAP-1 and RXMAP-2 occupy 53.11% and 53.31% lesser area compared to XTEA I and 52.97% and 53.18% lesser area compared to XTEA II implementation. The total power consumed by the proposed encoder architectures RXMAP-1 and RXMAP-2 is 68.76% and 71.64% lesser than XTEA II implementation, respectively, while maintaining the equal throughput.