Intrusion Detection Based on Bidirectional Long Short-Term Memory with Attention Mechanism

With the recent developments in the Internet of Things (IoT), the amount of data collected has expanded tremendously, resulting in a higher demand for data storage, computational capacity, and real-time processing capabilities. Cloud computing has traditionally played an important role in establishing IoT. However, fog computing has recently emerged as a new field complementing cloud computing due to its enhanced mobility, location awareness, heterogeneity, scalability, low latency, and geographic distribution. However, IoT networks are vulnerable to unwanted assaults because of their open and shared nature. As a result, various fog computing-based security models that protect IoT networks have been developed. A distributed architecture based on an intrusion detection system (IDS) ensures that a dynamic, scalable IoT environment with the ability to disperse centralized tasks to local fog nodes and which successfully detects advanced malicious threats is available. In this study, we examined the time-related aspects of network traffic data. We presented an intrusion detection model based on a two-layered bidirectional long short-term memory (Bi-LSTM) with an attention mechanism for traffic data classification verified on the UNSW-NB15 benchmark dataset. We showed that the suggested model outperformed numerous leading-edge Network IDS that used machine learning models in terms of accuracy, precision, recall and F1 score.     

Fractional Order Nonlinear Bone Remodeling Dynamics Using the Supervised Neural Network

This study aims to solve the nonlinear fractional-order mathematical model (FOMM) by using the normal and dysregulated bone remodeling of the myeloma bone disease (MBD). For the more precise performance of the model, fractional-order derivatives have been used to solve the disease model numerically. The FOMM is preliminarily designed to focus on the critical interactions between bone resorption or osteoclasts (OC) and bone formation or osteoblasts (OB). The connections of OC and OB are represented by a nonlinear differential system based on the cellular components, which depict stable fluctuation in the usual bone case and unstable fluctuation through the MBD. Untreated myeloma causes by increasing the OC and reducing the osteoblasts, resulting in net bone waste the tumor growth. The solutions of the FOMM will be provided by using the stochastic framework based on the Levenberg-Marquardt backpropagation (LVMBP) neural networks (NN), i.e., LVMBPNN. The mathematical performances of three variations of the fractional-order derivative based on the nonlinear disease model using the LVMPNN. The static structural performances are 82% for investigation and 9% for both learning and certification. The performances of the LVMBPNN are authenticated by using the results of the Adams-Bashforth-Moulton mechanism. To accomplish the capability, steadiness, accuracy, and ability of the LVMBPNN, the performances of the error histograms (EHs), mean square error (MSE), recurrence, and state transitions (STs) will be provided.     

Correction on link relative based approach to CUSUM chart

Link relative-based approach was used in an article (see reference 1) to enhance the performance of the cumulative sum (CUSUM) control chart. This technique involves the use of firstly, the link relative variable to convert the process observations in a relative to the mean form and secondly, optimal constants to define a new variable which is used as the plotting statistic of the link relative CUSUM chart. In this article, it is proven through simulation study that the optimal constants with fixed values, as reported in the aforementioned article, give different results. Instead, if the regression technique is used, then the same results will be obtained.     

Effect of carbide ceramics and coke on the properties of dispersion strengthened aluminium-silicon7-magnesium hybrid composites

In this study, aluminium-silicon7-magnesium alloy was reinforced with carbide ceramics (silicon carbide and boron carbide) and coke microparticles. The composite samples (S1, S2, and S3) with 4 %, 8 %, and 12 % by weight of silicon carbide and boron carbide microparticles and 8 % by weight of coke micro particles respectively, were prepared through the ultrasonic assisted melt-stir casting route. It was observed from the literature that only a few research works reported the cumulative effect of adding these micro-particulate reinforcements within aluminium-silicon7-magnesium alloy. The microstructural evaluation revealed uniform dispersion of particulate reinforcements. Area energy dispersive x-ray spectroscopy scan results proved the presence of the reinforcement particles as heterogeneous phases at distinct zones. The samples undergone tensile tests failed under the brittle fracture mode. The ultimate tensile strength of the composites improved by about 27.4 %, the % elongation improved by about 6.8 %, and Vickers microhardness dropped 1.9 % to the maximum. Dimples were observed in the fracture morphology analysis and it was mainly due to the mixed ductility of the sample S2. An abrasive wear mechanism was observed in the worn-out sample S2. The wear rate decreased significantly with an increase in weight fraction of the reinforcements.     

NRP/B, a novel nuclear matrix protein, associates with p110(RB) and is involved in neuronal differentiation

The nuclear matrix is defined as the insoluble framework of the nucleus and has been implicated in the regulation of gene expression, the cell cycle, and nuclear structural integrity via linkage to intermediate filaments of the cytoskeleton. We have discovered a novel nuclear matrix protein, NRP/B (nuclear restricted protein/brain), which contains two major structural elements: a BTB domain-like structure in the predicted NH2 terminus, and a "kelch motif" in the predicted COOH-terminal domain. NRP/B mRNA (5.5 kb) is predominantly expressed in human fetal and adult brain with minor expression in kidney and pancreas. During mouse embryogenesis, NRP/B mRNA expression is upregulated in the nervous system. The NRP/B protein is expressed in rat primary hippocampal neurons, but not in primary astrocytes. NRP/B expression was upregulated during the differentiation of murine Neuro 2A and human SH-SY5Y neuroblastoma cells. Overexpression of NRP/B in these cells augmented neuronal process formation. Treatment with antisense NRP/B oligodeoxynucleotides inhibited the neurite development of rat primary hippocampal neurons as well as the neuronal process formation during neuronal differentiation of PC-12 cells. Since the hypophosphorylated form of retinoblastoma protein (p110(RB)) is found to be associated with the nuclear matrix and overexpression of p110(RB) induces neuronal differentiation, we investigated whether NRP/B is associated with p110(RB). Both in vivo and in vitro experiments demonstrate that NRP/B can be phosphorylated and can bind to the functionally active hypophosphorylated form of the p110(RB) during neuronal differentiation of SH-SY5Y neuroblastoma cells induced by retinoic acid. Our studies indicate that NRP/B is a novel nuclear matrix protein, specifically expressed in primary neurons, that interacts with p110(RB) and participates in the regulation of neuronal process formation.     

State complexity of unique rational operations

For each basic language operation we define its “unique” counterpart as being the operation that results in a language whose words can be obtained uniquely through the given operation. These unique operations can arguably be viewed as combined basic operations, placing this work in the popular area of state complexity of combined operations on regular languages. We study the state complexity of unique rational operations and we provide upper bounds and empirical results meant to cast light into this matter. Equally important, we hope to have provided a generic methodology for estimating their state complexity.     

Recommendation engine based on derived wisdom for more similar item neighbors

Information Systems and e-Business Management - Collaborative filtering (CF) is a popular and widely accepted recommendation technique. CF is an automated form of word-of-mouth communication...     

Fruit Senescence and Decay Simulation

Aging and imperfections provide important visual cues for realism. We present a novel physically‐based approach for simulating the biological aging and decay process in fruits. This method simulates interactions between multiple processes. Our biologically‐derived, reaction‐diffusion model generates growth patterns for areas of fungal and bacterial infection. Fungal colony spread and propagation is affected by both bacterial growth and nutrient depletion. This process changes the physical properties of the surface of the fruit as well as its internal volume substrate. The fruit is physically simulated with parameters such as skin thickness and porosity, water content, flesh rigidity, ambient temperature, humidity, and proximity to other surfaces. Our model produces a simulation that closely mirrors the progression of decay in real fruits under similar parameterized conditions. Additionally, we provide a tool that allows artists to customize the input of the program to produce generalized fruit simulations.