Search

Abstract An interval type-2 (IT2) mutual subsethood Cauchy fuzzy neural inference system has been proposed in this paper. The network architecture consists of 3-layers with all connection weights being IT2 Cauchy fuzzy membership functions (CMFs). The crisp inputs to the system are fuzzified into IT2CMFs with fixed centers and uncertain spreads. The hidden layer represents the rule-based knowledge. The firing degree of the antecedent part of each rule at the hidden layer is computed by aggregating the product of the mutual subsethood similarity measures between the inputs and the connection weights. A volume defuzzification is used to compute the numeric output. A gradient descent back-propagation algorithm is used to train the model. The novelty of the proposed model is threefold. First, is enriching the theory of the mutual subsethood fuzzy neural models by adopting the Cauchy membership function (CMF) as another powerful fuzzy basis function (FBF) rather than the classical choice of Gaussian fuzzy membership functions (GMFs). Second, is the success of computing the mutual subsethood similarity measure between the IT2CMFs and all the model parameters’ updating equations in analytic closed-form formulas, not numerically or approximately. Third, is the ability to extract the type-1 (T1) mutual subsethood Cauchy fuzzy neural inference system (T1MSCFuNIS) with all its analytic closed-form formulas directly as a special case from the general formulas of IT2MSCFuNIS model. Such a novelty makes the proposed model a concrete and effective development of the theory of mutual subsethood fuzzy neural models. Both IT2MSCFuNIS and T1MSCFuNIS models have been tested using different examples from the domains of function approximation, classification, and prediction. The results ensure the efficacy of both models compared with other models reported in the literature.

Most populated developing countries having water resources, like Egypt, are interested in aquaculture since it supplies around 30% of the cheap protein consumed by customers. Increasing the production of aquaculture, specifically fish farming, in such countries represents an essential need. One candidate water resource for freshwater fish farming in Egypt is the Nile River (1530 km long). Yet, this represents a challenging task due to the existing variations in its water quality (WQ) parameters, such as dissolved oxygen, acidity, and temperature, at different sites. Climate change and pollution negatively affect many water quality parameters. This work provides a fuzzy-based approach for modeling WQ requirements for a set of fish types and evaluates the suitability of a water site for farming them. Thus, it greatly helps managing and planning fish farming in a set of water sites. It benefits from the flexibility of fuzzy logic to model the farming requirements of each fish type. Consequently, it evaluates and clusters the water sites with respect to their degrees of suitability for farming various fish types. The illustrative case study considers 27 freshwater sites spread along the Nile River and 17 freshwater fish types. The result incorporates a set of suitable clusters and a set of unsuitable ones for farming each fish type. It greatly helps managing and planning fish farming, to maximize the overall productivity and prevent probable catastrophic damage. In addition, it shows how to enhance each unsuitable site. We believe that eliminating the causes of pollution in the polluted freshwater sites along a water source could cause a significant boom in the cultivation of multiple freshwater fish types.

Shill attacks are a serious threat to the stability of filtering and recommendation systems. These attacks involve the injection of fake profiles into the system, which can compromise the reliability of system output. Several shilling attack detection techniques have been proposed, but they often have limitations in terms of accuracy. This works presents a enhanced method for detecting attacks in collaborative recommender systems. The proposed method is based on a combination of statistical and machine learning techniques. The statistical techniques are used to identify anomalous user behavior, while the machine learning techniques are used to classify users as either malicious or benign. The main contribution of the proposed method is the use of a hybrid approach that combines the strengths of statistical and machine learning techniques. The statistical techniques are able to identify anomalous user behavior that is not easily detected by machine learning techniques. The machine learning techniques are able to classify users as either malicious or benign with a high degree of accuracy. The proposed method was evaluated on a real-world dataset. The results showed that the proposed method was able to detect attacks with a high degree of accuracy. The proposed method uses a combination of ensemble learning and feature selection to achieve better accuracy than previous methods. The results of the experiments show that the proposed method can achieve an accuracy of up to 99%.

The stability and reliability of filtration and recommender systems are crucial for continuous operation. The presence of fake profiles, known as “shilling attacks,” can undermine the reliability of these systems. Therefore, it is important to detect and classify these attacks. Numerous techniques for detecting shilling attacks have been proposed, including supervised, semi-supervised, unsupervised, Deep Learning, and hyper deep learning methods. These techniques utilize well-known shilling attack models to target collaborative recommender systems. While previous research has focused on evaluating shilling attack strategies from a global perspective, considering factors such as attack size and attacker’s knowledge, there is a lack of comparative studies on the various existing and commonly used attack detection methods. This paper aims to fill this gap by providing a comprehensive survey of shilling attack models, detection attributes, and detection algorithms. Furthermore, we explore the traits of injected profiles that are exploited by detection algorithms, which has not been thoroughly investigated in prior works. We also conduct experimental studies on popular attack detection methods. Our experimental results reveal that hybrid deep learning algorithms exhibit the highest performance in shilling detection, followed by supervised learning algorithms and semi-supervised learning algorithms. In contrast, the unsupervised technique performs poorly. The deep learning-based Shilling Attack Detection demonstrates accuracy and quality in accurately identifying a variety of mixed attacks. This study provides valuable insights into shilling attack models, detection attributes, and detection algorithms. Our findings highlight the superior performance of hybrid deep learning algorithms in shilling detection, as well as the limitations of unsupervised techniques. Deep learning-based Shilling Attack Detection showcases its effectiveness and accuracy in identifying various types of attacks.

In artificial neural networks, activation functions play a significant role in the learning process. Choosing the proper activation function is a major factor in achieving a successful learning performance. Many activation functions are sufficient universal approximators, but their performance is lacking. Thus, many efforts have been directed toward activation functions to improve the learning performance of artificial neural networks. However, the learning process involves many challenges, such as saturation, dying, and exploding/vanishing the gradient problems. The contribution of this work resides in several axes. First, we introduce two novel activation functions: absolute linear units and inverse polynomial linear units. Both activation functions are augmented by an adjustable parameter that controls the slope of the gradient. Second, we present a comprehensive study and a taxonomy of various types of activation functions. Third, we conduct a broad range of experiments on several deep neural architecture models with consideration of network type and depth. Fourth, we evaluate the proposed activation functions’ performance in image and text classification tasks. For this purpose, several public benchmark datasets are utilized to evaluate and compare the performance of the proposed functions with that of a group of common activation functions. Finally, we deeply analyze the impact of several common activation functions on deep network architectures. Results reveal that the proposed functions outperform most of the popular activation functions in several benchmarks. The statistical study of the overall experiments on both classification categories indicates that the proposed activation functions are robust and superior among all the competitive activation functions in terms of average accuracy.

The airport ground handling has a global trend to meet the Service Level Agreement (SLA) requirementsthat represents resource allocation with more restrictions according to flights. That can be achieved by predictingfuture resources demands. this research presents a comparison between the most used machine learning techniquesimplemented in many different fields for demand prediction and resource allocation. The prediction model nomi-nated and used in this research is the Support Vector Machine (SVM) to predict the required resources for eachflight, despite the restrictions imposed by airlines when contracting their services in the Service Level Agreement.The approach has been trained and tested using real data from Cairo International Airport. the proposed (SVM)technique implemented and explained with a varying accuracy of resource allocation prediction, showing thateven for variations accuracy in resource prediction in different scenarios; the Support Vector Machine techniquecan produce a good performance as resource allocation in the airport.

Many approaches were presented recently for developing Textual Case-Based Reasoning (TCBR) applications. One of the successful approaches is SOPHisticated Information Analysis (SOPHIA), which is distinguished by its ability to work without prior knowledge engineering, without domain dependency and without language dependency. SOPHIA is based on the distributional document clustering approach, which facilitates an advanced and rich knowledge discovery framework for case-based retrieval. This paper contributes to propose enhancements to SOPHIA approach that aims to enhance the retrieval efficiency and increase the precision degree. It also aimed to grantee that all results will have the same subject of the user query. The enhancements include performing an automatic classification to the case-base before the clustering step in the indexing stage, and include performing an automatic classification to the user query before the retrieval stage. Moreover, proofing that SOPHIA approach is a domain and language independent by applying it in the domain of Islamic jurisprudence in Arabic language.

In the analytic network process (ANP) a hierarchy or network is created to represent a decision and establishes a matrix containing the pair wise comparison judgments for the elements linked under a parent element. A priority vector of relative weights for these elements is derived. Then all the priority vectors are appropriately weighted and summed to obtain the overall priorities for the alternatives of a decision. In this paper we will develop an efficient fuzzy ANP model which helps decision makers to choose among the alternatives for the Egyptian scenarios of electrical power generation.

Dynamic routing algorithms play an important role in road traffic routing to avoid congestion and to direct vehicles to better routes. AntNet routing algorithms have been applied, extensively and successfully, in data communication network. However, its application for dynamic routing on road networks is still considerably limited. This paper presents a modified version of the AntNet routing algorithm, called “Threshold based AntNet”, that has the ability to efficiently utilize a priori information of dynamic traffic routing, especially, for road networks. The modification exploits the practical and pre-known information for most road traffic networks, namely, the good travel times between sources and destinations. The values of those good travel times are manipulated as threshold values. This approach has proven to conserve tracking of good routes. According to the dynamic nature of the problem, the presented approach guards the agility of rediscovering a good route. Attaining the thresholds (good reported travel times), of a given source to destination route, permits for a better utilization of the computational resources, that, leads to better accommodation for the network changes. The presented algorithm introduces a new type of ants called “check ants”. It assists in preserving good routes and, better yet, exposes and discards the degraded ones. The threshold AntNet algorithm presents a new strategy for updating the routing information, supported by the backward ants.

Web-Based Applications (WBA) are fast becoming more widespread, larger, more interactive, and more essential to the international use of computers. The most successful WBA companies are beginning to realize that key critical factors of success or failure of any WBA must be highly dependable on delivering on a high quality web site. To attain the desired quality of WBA, it is necessary to suggest a model that organizes and enables the identification of WBA quality perspectives. This paper addresses WBA quality model and categorizes its quality factors. The software is an essential part of any WBA. ISO9126 standard for software engineering product quality states that the main purpose of software quality evaluation is to provide quantitative reference for software products evaluation that is reliable, understandable, and acceptable. The main weakness point here is the lack of a formal specification of key factors for WBA quality. Traditional quality models are not adequate for WBA because they do not address all problems associated with the new features of WBA. Therefore, ISO9126 and different quality models of software were investigated and partially used as an initial step to identify a conceptual quality model for WBA. WBA have common characteristics with traditional software packages, and other distinct characteristics that are particular to WBA. In this paper a proposed conceptual quality model to organize WBA quality factors in terms of its sub factors was identified. In addition, the proposed conceptual quality model effectively reflects the main views of WBA based on the opinion of highly skilled professionals (visitor, owner, end user). The main goal of this paper is identifying, categorizing, and modeling WBA quality factors.

Abstract Background Activation of the platelet plays an important role in the process of atherosclerosis. Mean platelet volume (MPV) is significantly associated with the poor outcome of acute ischemic stroke while the results of studies about the relationship between plateletcrit (PCT) and stroke outcome were inconsistent. The aim of this work is to determine whether an association exists between MPV and plateletcrit (PCT) and outcome of acute ischemic stroke. Methods We examined 157 patients with ischemic stroke, admitted to the Sohag University Hospital. The diagnosis of stroke was performed clinically according to The World Health Organization and confirmed by brain CT and MRI when needed. Platelet indices including MPV and PCT were assessed immediately (within 2 h) after admission. After 3 months, the functional outcome was assessed using the modified Rankin Scale (mRS) with assessment of the relationship between platelet indices and stroke outcome. Results About 50% of the participants have favorable outcome. MPV was significantly higher in the unfavorable group (10.4 ± 2.3 fL) than in the favorable one (8.7 ± 1.3 fL) (P