Recently, robot arms have become an irreplaceable production tool, which play an important role in the industrial production. It is necessary to ensure the absolute positioning accuracy of the robot to realize automatic production. Due to the influence of machining tolerance, assembly tolerance, the robot positioning accuracy is poor. Therefore, in order to enable the precise operation of the robot, it is necessary to calibrate the robotic kinematic parameters. The least square method and Levenberg-Marquardt (LM) algorithm are commonly used to identify the positioning error of robot. However, it generally has the overfitting caused by improper regularization schemes. To solve this problem, this article discusses six regularization schemes based on its error models, i.e., $L_{1}$ , $L_{2}$ , dropout, elastic, log, and swish. Moreover, this article proposes a scheme with six regularization to obtain a reliable ensemble, which can effectively avoid overfitting. The positioning accuracy of the robot is improved significantly after calibration by enough experiments, which verifies the feasibility of the proposed method.

DDoS (Distributed Denial of Service) attacks have now become a serious risk to the integrity and confidentiality of computer networks and systems, which are essential assets in today’s world. Detecting DDoS attacks is a difficult task that must be accomplished before any mitigation strategies can be used. The identification of DDoS attacks has already been successfully implemented using machine learning/deep learning (ML/DL). However, due to an inherent limitation of ML/DL frameworks—so-called optimal feature selection—complete accomplishment is likewise out of reach. This is a case in which a machine learning/deep learning-based system does not produce promising results for identifying DDoS attacks. At the moment, existing research on forecasting DDoS attacks has yielded a variety of unexpected predictions utilising machine learning (ML) classifiers and conventional approaches for feature encoding. These previous efforts also made use of deep neural networks to extract features without having to maintain the track of the sequence information. The current work suggests predicting DDoS attacks using a hybrid deep learning (DL) model, namely a CNN with BiLSTM (bidirectional long/short-term memory), in order to effectively anticipate DDoS attacks using benchmark data. By ranking and choosing features that scored the highest in the provided data set, only the most pertinent features were picked. Experiment findings demonstrate that the proposed CNN-BI-LSTM attained an accuracy of up to 94.52 percent using the data set CIC-DDoS2019 during training, testing, and validation.

Internet of Things (IoT) is promising technology that brings tremendous benefits if used optimally. At the same time, it has resulted in an increase in cybersecurity risks due to the lack of security for IoT devices. IoT botnets, for instance, have become a critical threat; however, systematic and comprehensive studies analyzing the importance of botnet detection methods are limited in the IoT environment. Thus, this study aimed to identify, assess and provide a thoroughly review of experimental works on the research relevant to the detection of IoT botnets. To accomplish this goal, a systematic literature review (SLR), an effective method, was applied for gathering and critically reviewing research papers. This work employed three research questions on the detection methods used to detect IoT botnets, the botnet phases and the different malicious activity scenarios. The authors analyzed the nominated research and the key methods related to them. The detection methods have been classified based on the techniques used, and the authors investigated the botnet phases during which detection is accomplished. This research procedure was used to create a source of foundational knowledge of IoT botnet detection methods. As a result of this study, the authors analyzed the current research gaps and suggest future research directions.

Blockchain technology has wide applications in the fields of finance, public welfare, and the Internet of Things. Owing to a blockchain's characteristics, which include decentralization, openness, autonomy, immutability, and anonymity, it is difficult to quickly reach a reliable consensus result among its nodes. This work proposes a performance-optimized consensus mechanism based on node classification. Nodes are classified into accounting, validating, and propagating ones based on their trust values. All accounting nodes form an accounting node group, from which one is selected as the current accounting node to package transactions into a block, and the remaining nodes in the accounting node group can be used to validate the block quickly, owing to their high trust values. Validating and propagating nodes are responsible for validating and propagating transactions, respectively. All nodes' trust values are dynamically updated according to their behavior and performance. Corresponding algorithms are designed to realize the proposed consensus mechanism. The experimental results show that the proposed consensus mechanism provides higher throughput, lower consumption, and higher fault tolerance than some popularly used methods, thereby advancing the field of consortium blockchains.

As the amount of historical data available in the legal arena has grown over time, industry specialists are driven to gather, compile, and analyze this data in order to forecast court case rulings. However, predicting and justifying court rulings while using judicial facts is no easy task. Currently, previous research on forecasting court outcomes using small experimental datasets yielded a number of unanticipated predictions utilizing machine learning (ML) models and conventional methodologies for categorical feature encoding. The current work proposes forecasting court judgments using a hybrid neural network model, namely a long short-term memory (LSTM) network with a CNN, in order to effectively forecast court rulings using historic judicial datasets. By prioritizing and choosing features that scored the highest in the provided legal data set, only the most pertinent features were picked. After that, the LSTM+CNN model was utilized to forecast lawsuit verdicts. In contrast to previous related experiments, this composite model’s testing results were promising, showing 92.05 percent accuracy, 93 percent precision, 94 percent recall, and a 93 percent F1-score.

Wireless sensor networking is being used extensively in agricultural activities to increase productivity and reduce losses in various ways. The greenhouse simplifies the concept of planting, which has several benefits in agriculture. In agricultural models, soil pH sensors and gas sensors are commonly used. These sensors are applicable in various Internet of Things (IoT) integrated agricultural activities. The paper discusses the hardware design and working of the proposed model. In addition, various agricultural models used for evapotranspiration are also explained. The key factors such as congestion control are evaluated using the Penman-Monteith equation. This paper focuses on implementing more than two references parameters like evapotranspiration and humidity under different conditions, which aids in splitting the relationship evenly by the number of sources. Furthermore, the paper shows the implementation done with MATLAB and values are adjusted using the code. The paper claims to achieve similar variations with the same source value, validating the proposed model’s efficiency and fairness. In an optimal region, these schemes also demonstrate higher throughput and lower delay rates. The improved packet propagation through the IoT network is demonstrated using visualization tools, and the feedback is computed to determine the overall access amount (A1 + A2) obtained. The experimental results show that the propagation rate is 1.24, more significant than the link capacity value. The claims are verified by showing the improved congestion control as it outperforms different parameters, considering an additive increase condition by 0.3% and multiplicative decrease condition by 1.2 %.