Clustered Wireless Sensor Networks (WSNs) help in the construction of robust and scalable network infrastructure which increases the probability of minimizing energy consumption with extended network lifetime. But the clustered WSNs pose the challenges of non-uniform energy consumption, inadequate cluster head allocation and imbalanced distribution of load in the network. This challenges dramatically impact the network lifetime when improper clusters are constructed. This improper clusters in turn makes the sensor node to prematurely die due to increased energy consumption. Potential cluster formation and optimal cluster head selection techniques are essential for the purpose of improving the clustering quality that contributes towards better energy stability and extended network lifetime. In this paper, Boosted Sooty Tern Optimization Algorithm-based protocol with multiple objectives (BSHPFMOCS) is proposed for enhancing the quality of clustering with the objective of improving energy stability and prolonged network lifetime in clustered WSNs. This BSHOA facilitates an accurate search process which helps in selecting optimal CHs depending on the fitness function that concentrates on the improvement of clusters’ aggregation. This clustering protocol incorporated an advanced cluster formation strategy which entrusted the CHs to select their own cluster members depending on minimized intra-cluster distance. It further included Piranhav Foraging Optimization Algorithm (PFOA) for employing sink mobility that addresses the problem of hot-spot in WSNs. The simulation results of BSHOA protocol confirmed better network lifetime of 10.76

Social networks serve as dependable platforms for valuable communication channels and the dissemination of information on a global scale, with millions of users utilizing them on a daily basis. While they have emerged as a means of disseminating information, they have also quickly transformed into a conduit for spreading false information, rumors, unsolicited messages, promotional content, fabricated news, and other undesirable content. Both spammers and non-spammers are present on social networks, with spammers typically sending advertisements and unwanted messages to users. Unfortunately, many users unknowingly spread unwanted messages, contributing to the problem. To address this issue, a model has been developed to recognize spammers on social networks. This process is based on the user's actions and patterns, as well as information-based features such as URLs, posts, replies, mentions, tweets, and retweets. The Recursive Feature Elimination (RFE) method is used in conjunction with five classifiers, including Random Forest, Logistic Regression, SVM, XGB, and Adaptive Boosting. Data is collected from social networks and implemented using the Sci-Kit Learn library in Python. The model's performance is evaluated using metrics such as TP and FP rate, precision, and accuracy.

A cognitive state (CS) assessment can be effectively performed using Electroencephalogram (EEG). However, due to the curse of dimensionality issues of EEG, most of the clustering methods often lead to poor performance. Deep neural network-based representation learning transforms high-dimensional data into lower-dimensional feature space, increasing the performance of CS prediction in students. So, in this research, a novel graph-based conventional attention neural network (GA-CNN) is developed to reduce distribution differences while analyzing student performance. To perform this classification, the obtained EEG signal dataset is denoised with the help of Stationary Wavelet Transform (SWT) based Independent Component Analysis (ICA) that filters out the noise signals. Likewise, the behavioural dataset is pre-processed using zero normalization and Not a Number (NaN) value computation method. Then, the effective features are extracted from the pre-processed data using the Long-Short-Term-Memory (LSTM) technique. Finally, the GA-CNN model is initiated to classify the students' cognitive state (CS). The proposed method is implemented using the MATLAB tool, and the performance of the expected GA-CNN model is compared to other approaches where the analysis is done using the benchmark Kaggle EEG data set. The classification accuracy is significantly improved compared to other methods. The model achieves 87% accuracy, 98% precision, 75% recall, and 85% F1 score, outperforming various methods and making a better compromise.

Natural language processing innovations in the past few decades have made it feasible to synthesis and comprehend coherent text in a variety of ways, turning theoretical techniques into practical implementations. Both report summarizing software and sectors like content writers have been significantly impacted by the extensive Language-model. A huge language model, however, could show evidence of social prejudice, giving moral as well as environmental hazards from negligence, according to observations. Therefore, it is necessary to develop comprehensive guidelines for responsible LLM (Large Language Models). Despite the fact that numerous empirical investigations show that sophisticated large language models has very few ethical difficulties, there isn't a thorough investigation and consumers study of the legality of present large language model use. We use a qualitative study method on OpenAI's ChatGPT3 to solution-focus the real-world ethical risks in current large language models in order to further guide ongoing efforts on responsibly constructing ethical large language models. We carefully review ChatGPT3 from the four perspectives of bias and robustness. According to our stated opinions, we objectively benchmark ChatGPT3 on a number of sample datasets. In this work, it was found that a substantial fraction of principled problems are not solved by the current benchmarks; therefore new case examples were provided to support this. Additionally discussed were the importance of the findings regarding ChatGPT3's AI ethics, potential problems in the future, and helpful design considerations for big language models. This study may provide some guidance for future investigations into and mitigation of the ethical risks offered by technology in large Language Models applications.

Ensuring the safety of underground miners is of utmost importance in the mining industry, given the devastating consequences of mining accidents. To address this concern, this research study has developed an intelligent headgear equipped with an array of sensors to continuously monitor the working conditions of miners. This advanced device is engineered to identify critical parameters such as temperature, humidity, concentrations of harmful gases, and vibrations. It is equipped with dedicated sensors for each of these parameters, and it promptly transmits real-time data to a centralized control centre. The control centre can analyse this information and take immediate actions, which may include worker evacuation or improvements to working conditions. The proposed system utilizes an Arduino microcontroller for processing the data from these sensors and then transmits it via a Wi-Fi module. The proposed model has utilized Cloud-based loT and machine learning algorithms to enhance the accuracy of the sensor readings. Specifically, an Artificial Neural Network (ANN) algorithm is employed to train a dataset of sensor information stored in CS V format. This innovative system has the potential to significantly elevate safety standards and minimize the risk of mining accidents.

Companies like Ola, Uber, and others work in a similar way which is they hire a third-party vehicle and driver i.e. locals of any particular area. These drivers will be connected with the customers who want to avail the taxi service. This makes it easy for the drivers to find their customers faster and also helps people to book any cab at the last minute. This last-minute booking was the one thing that made these applications succeed in their business. Our idea is to enhance their business by getting additional information from the third-party vehicles like petrol level, the heat level of the engines, and also detect accidents that occurred. We need to do this because there are accidents that occur due to driver's negligence. It is the sole duty of the driver to check these things in the car or any vehicle before even starting or getting onto the road but some drivers are not checking their vehicles properly and some vehicles are not even serviced regularly driving those vehicles can create unnecessary problems both to the drivers and customers. This leaves a bad impression on the company that provides this service and affects its reputation. Now to find these parameters in the car that also in real-time we will be using IoT, also called the IoT module or IoT chip is a small electronic device that is embedded on objects or any machine used to send or receive signals when connected to the internet. The IoT chip provides always-on connectivity because it transfers signals in real-time such that there is no need to switch it on every time to use it. These details will then be sent to the application that will be used by Android users. In the application, all the parameters like temperature and voltage will be acquired using sensors like (Voltage sensor ZMPT101B, Temperature Sensor LM35, MEMS sensor ADXL345) and the acquired values will be available. Later this will be integrated into the respective company's application so that the customers can also easily see the vehicle they are travelling to and in case of danger they can warn the driver.

Energy efficient routing has been the mainstay of research in wireless sensor networks. With rising utility of WSNs in a broad range of applications, a significant volume of research contributions are targeted towards obtaining an energy efficient routing with a good quality of service (QoS) metric. In recent times, the utility of evolutionary algorithm for such multi-objective problems has been on the rising trend as evolutionary models closely mimic the real time scenario of any problem optimization. In this research work, a modified ant colony optimization (M-ACO) has been proposed and implemented for optimization in terms of energy conservation and thereby to extend the network lifetime of the network

The world's medical industry has faced many significant challenges due to the COVID-19 pandemic, which occurred in 2019. Accurate and speedy detection procedures are essential requirements for identifying and isolating infected individuals, thereby reducing the further spread of viruses within groups. This would help to effectively control the spread of viruses. Artificial intelligence-powered deep learning models have emerged as a promising approach for COVID-19 identification. These models use massive volumes of medical data, such as X-ray and CT scan images, to identify patterns and characteristics associated with the infection. Deep learning algorithms can detect small irregularities by training on varied datasets, allowing for early diagnosis and treatments. By giving medical professionals the necessary resources for efficient screening, deep learning has the ability to completely transform COVID-19 identification and management. Deep learning models can enhance patient outcomes, optimize resource allocation, and lessen the burden on healthcare systems amid this global health crisis because of their capacity to assess complex medical data and support decision-making. This research paper presents a DL-based method for identifying COVID-19 cases from chest X-ray images. The images of normal chest and affected images are collected in real-time and used to build a convolutional neural network (CNN) model. Through effective differentiation between COVID-19 and normal instances, the model offers efficient screening. Experimental results illustrate the utility and real-world applicability of the model. The proposed technology not only helps medical personnel identify and diagnose COVID-19 early on, but it also promotes public health activities.

The covid 19 pandemic has made everything virtual, including education. It is difficult to tell if students are focused or not due to online education. To help teachers, we are developing a framework for recognizing and assessing student focus. By using the concept of brain-computer communication, we can find the student's concentration level. The data obtained from the electroencephalogram (EEG) signals is used as a data set to predict concentration levels. A four-channel device is used to capture brain waves. The data were preprocessed and feature extraction was performed to determine the concentration level as active or inactive. In this method, we use a multiclass approach to develop a deep learning model that uses LSTM to classify concentration into low, or high concentration levels with accuracy of 88%.

Agriculture is one of the most important assets of the Indian economy, as well as the primary occupation of a large number of people. In comparison to traditional farming, Internet of Things (IoT)-based farming is extremely environmentally friendly. Root Rot can be detected early with the proposed approach. Root rot can be caused by a variety of organisms, including bacteria and fungus. Farmers, on the other hand, generally refer to all forms of bad root browning as root rot. Farming innovation isn't new, but the Internet of Things is going to propel smart farming to the next level. The Internet of Things is a method that uses actuators or sensors to provide direct or indirect access to the internet. The proposed approach has a number of features, including root rot detection, serving as a primary far-off tracking machine, humidity and temperature sensing, and nutrient and pH stage of soil detection. Varioussensors, such as the DHT11 sensor, optical transducer, and pH sensor, are deployed at various locations around farms, and microcontrollers are used to control many of these sensors. The sensor data is delivered to UbiDots, a cloud-based storage platform, where it is processed and a notification about farm conditions is sent to the user.

Liver disease can be genetic. The liver can be damaged by alcohol use, obesity and hepatitis viruses. Chronic liver disease has many causes like Liver cancer, Cirrhosis and life threatening conditions. So prevention treatment can help to heal the damage and prevent liver failure. Hence, the time required to introduce the machine learning techniques which minimize the error and give more accuracy. Aim of this research paper is categories into two phases. In the first phase six classification algorithms (LR, KNN, SVM, NB, DT, and RF) are compared with ML Classification evaluation metrics. As a consequence, Logistic Regression obtained a higher accuracy of 71.6% in liver disease prediction than the other classification algorithms. In Phase 2, we selected a sample from a classification method with lower accuracy, such as Random Forest, Decision Tree algorithms and we adjusted the hyper parameters. RF accuracy is improved after Hyper Parameter Optimization by +4%, whereas Decision Tree algorithm accuracy is improved by +12%.

In today's world, women's safety is a common concern, and it is everyone's duty to address it. This paper describes a “GPS and GSM Circuit based lady's security framework with an alarm and a shock generating device,” as well as a “GPS and GSM Circuit based ladies security framework with an alarm and a shock generating system.” This offers a collection of GPS devices that are primarily used to monitor the location [1]. Aside from that, the device will send out warnings, electric shocks, and crisis capture triggers. The main goal of this paper is to show a device that is so conservative that it provides a desirable disguised role. The main feature of this platform is that it does not include the use of a smart phone, which sets it apart from other applications currently available. The device is easy to use and blends in with the proposed apparatus. The device is resistant to all weather conditions and can examine any possibility of assisting the trick in every kind of emergency [3].

Agricultural productivity depends heavily on the economy. This stands as the main reason for the detection of the plant disease in the field of agriculture, because diseases in plants are quite natural. Crop diseases serve as a major food supply threat. The identification of diseases can lead to more rapid interventions in order to reduce the effects of plant disease. Automatic detection of plant leaf disease is beneficial because of the limited requirement of overviewing in large areas of crops and detects the diseases at an earliest, i.e., when they attack the plant leaves. The performance and accuracy level were quite low in existing system. Modern developments in deep learning have made drastic improvements in accuracy of object recognition. The proposed study aims at classifying the defect in the leaves of different plants using the plant leaf image. This is achieved using Faster Region based convolution neural network. The proposed method has the potential to identify various diseases possible in the plants.

As the implementation of sensing devices for healthcare increases, there has been much advancement and research work done because of its wide applications. There are number of devices that are connected to Internet of things that are used in many sensing support healthcare applications. However, transmitting and collecting data in a safe and secure manner is quite a challenge as it is prone to be attacked by hackers to access the data in an illegal manner. The solutions that are already used have issues like energy overheads, communication and storage problems. In order to overcome this issue, this research work proposes a FoG assisted device that can be used for secure and efficient management of the healthcare data. In the proposed methodology, a peer-to-peer communication means can be used to share the data with aggregate nodes which will then further communicate it with the FoG server. The FoG server is incorporated with the proposed algorithm in order to extract data from the aggressor node and further data is split to obtain device level data. Simulation results indicate that this scheme will help us attain 55% data size reduction when compared with the other methodologies. The results of this methodology observe significant improvement in terms of resilience, energy consumption, transmission ratio, communication and storage.

Data tampering and fraud in land records have increased drastically in the modern world. A data storage model using Blockchain and Interplanetary File System (IPFS) is proposed in this work. Land records and the farmer’s information are stored inside the Interplanetary file system. To avoid data faking, the hash address of the respective data generated by IPFS is stored in the blockchain. This proposed system when deployed on a large scale can outperform the existing methods of securing user data. One of the latest technological advancements in the software industry is the innovation of Blockchain Technology. This new technology has opened up a new business relationship platform that delivers feasibility, protection, and cheap rates. It provides a new foundation of trust for transactions that can facilitate a very streamlined workflow and a faster economy.

The teaching-learning process is seeing a big transformation in this digital age. It involves digital classrooms with various accessories of online tools such as video conferencing, digital materials, and other platforms for learning and assessment with options for both real-time and self-paced work in addition to the availability of teachers over video conferencing, text, phone, email, etc. To improve the online learning efficiency, assessing the cognitive state during the learning phase is highly required for the success of these developments. This work focused on cognitive state analysis during different learning tasks is determined by EEG brain signals that are captured using 128 channels Emotive Epoch headset device. Artifacts prominent in raw signals are filtered by linear filtering. Feature extraction for determination of concentration levels is done by applying fuzzy fractal dimension measures and Discrete Wavelet Transform (DWT) on the processed signals. The classification of extracted parameters into concentration levels is done by using deep learning algorithms like Enhanced Convolutional Neural Network (ECNN). This ECNN deep learning classification is highly accurate amongst all other remaining classifiers and is used as a feedback model to regulate this cognitive state.

Individuals support the incredible intensity of distributed computing, however can't completely believe the cloud suppliers to have protection delicate information, because of the nonappearance of client to-cloud controllability. To guarantee privacy, information proprietors re-appropriate scrambled information rather than plaintexts. To impart the scrambled documents to different clients, Cipher text-Policy Attribute-based Encryption (CP-ABE) can be used to direct fine-grained and proprietor driven access control. This is accomplished by keeping key position framework and capacity hubs in two unique ways. Over an unreliable channel, an open key is produced alongside the comparing private key and give to number of clients independently. The Key gave is free of different keys for every clients.

Online tools and platforms for assessment and learning are gaining importance in all ages of life. Stress monitoring is analyzed by measuring the cognitive states during learning and software development activities. Constant use of digital tools and medium for learning and development among end users leads to stress that gradually reduces the cognition outcomes. Short and long term stress factors leads to acute and chronic effects. Mental stress is monitored by acquired EEG signal that are the vital tools involved in the measurement of cognitive task like recall (question and answering), reading and coding skills. The signals captured using Emotiv Epoc Headset device with 14 channels is being used for measurement of mental stress. DWT and machine learning algorithms with enhanced CNN is implemented for feature extraction and stress (i.e High and Low) classification. Highly accurate feedback models are used to predict stress levels.

The use of pesticides, steroids and fertilizers has tremendously increased the negative effects caused to the people in terms of health. Harmful pesticides enter into the human body through fruits and vegetables, so an optimal solution is needed to recognize the disease and the pesticides in the fruits the common man is consuming. Hardware and a software design are done to obtain an accurate and a real time output. In this paper, a prototype of the system is developed with the use of four sensors, (temperature, gas, pH and moisture), Arduino microcontroller and a Wi-Fi module to get the information about the presence of pesticides. The maximum level of pesticides that is accepted legally to be consumed by animals and humans is given by MRL. If a fruit is detected to belong in a range above or below the MRL then it is said to contain pesticides. Through IoT, the pesticide content and the values obtained from each sensors are stored in the Cloud server MATLAB ThikSpeak. Coming to the software design, CNN and SVM algorithms are chosen and the image of the fruit is diagnosed by them. Two algorithms are mainly used to compare the accuracy produced by both and to select the most accurate between the two. Deep Learning process is performed on the image of the fruit and the disease affected in the fruit is identified and later stored in the Cloud server. The information about the disease in fruits and the pesticide value in fruits, the harmful effect caused by it, are sent to the cloud, which is then processed and sent to the application present in the consumer's smart phone which is developed in HTML5, thereby a real time regular monitoring is possible.