Ali Abdullah Amer

The fuzzy k-nearest neighbor comes to tackle the voting problem of the standard kNN, in which the same weight is assigned to each labeled sample, leading to a detrimental impact on the kNN performance. In fuzzy kNN, each instance’s fuzzy membership is found to produce seamless boundaries between classes. However, the computation of the memberships comes at additional costs due to the runtime overhead and memory requirements. Further, state-of-the-art fuzzy kNNs’ performance still suffers degradation because of class imbalance and outliers. To mitigate the impact of both problems, this study therefore develops two novel fuzzy models: Local Mean and Adaptive Learning fuzzy kNN (LMAL-FkNN) and Local Mean and Global Learning Fuzzy kNN (LMGL-FkNN). This is done by combining local mean vectors with class-based means and calculating the average local and global linkages in LMAl-FkNN and LMGL-FkNN, respectively. By calculating the local and class-based means of the global and local neighbors and using these means to compute the distance to the query, the impact of class imbalance is significantly lessened. Further, by incorporating class-based neighbors and using their means to find the final membership degrees, the outlier effects are substantially reduced. This dual approach enhances the robustness of the proposed models and thus improves the overall performance. To demonstrate the models’ competitiveness, a thorough evaluation study in five experimental phases is conducted against five state-of-the-art kNN rivals using forty-seven datasets. The results show that the LMGL-FkNN, in particular, has far more potential than its competitors over the vast majority of datasets.

Generative artificial intelligence (GenAI) has been developing with many incredible achievements like ChatGPT and Bard. Deep generative model (DGM) is a branch of GenAI, which is preeminent in generating raster data such as image and sound due to strong points of deep neural network (DNN) in inference and recognition. The built-in inference mechanism of DNN, which simulates and aims to synaptic plasticity of human neuron network, fosters generation ability of DGM which produces surprised results with support of statistical flexibility. Two popular approaches in DGM are Variational Autoencoders (VAE) and Generative Adversarial Network (GAN). Both VAE and GAN have their own strong points although they share and imply underline theory of statistics as well as incredible complex via hidden layers of DNN when DNN becomes effective encoding/decoding functions without concrete specifications. In this research, VAE and GAN is unified into a consistent and consolidated model called Adversarial Variational Autoencoders (AVA) in which VAE and GAN complement each other, for instance, VAE is good at generator by encoding data via excellent ideology of Kullback-Leibler divergence and GAN is a significantly important method to assess reliability of data which is realistic or fake. In other words, AVA aims to improve accuracy of generative models, besides AVA extends function of simple generative models. In methodology this research focuses on combination of applied mathematical concepts and skillful techniques of computer programming in order to implement and solve complicated problems as simply as possible.

In this work, a simple yet robust neighboring-aware hierarchical-based clustering approach (NHC) is developed. NHC employs its dynamic technique to take into account the surroundings of each point when clustering, making it extremely competitive. NHC offers a straightforward design and reliable clustering. It comprises two key techniques, namely, neighboring- aware and filtering and merging. While the proposed neighboring-aware technique helps find the most coherent clusters, filtering and merging help reach the desired number of clusters during the clustering process. The NHC's performance, which includes all evaluation metrics and run time, has been thoroughly tested against nine clustering rivals using four similarity measures on several real-world numerical and textual datasets. The evaluation is done in two phases. First, we compare NHC to three common clustering methods and show its efficacy through empirical analysis. Second, a comparison with six relevant, contemporary competitors highlights NHC's extremely competitive performance.

Distributed database system (DDBS) design is still an open challenge even after decades of research, especially in a dynamic network setting. Hence, to meet the demands of high-speed data gathering and for the management and preservation of huge systems, it is important to construct a distributed database for real-time data storage. Incidentally, some fragmentation schemes, such as horizontal, vertical, and hybrid, are widely used for DDBS design. At the same time, data allocation could not be done without first physically fragmenting the data because the fragmentation process is the foundation of the DDBS design. Extensive research have been conducted to develop effective solutions for DDBS design problems. But the great majority of them barely consider the RDDBS's initial design. Therefore, this work aims at proposing a clustering-based horizontal fragmentation and allocation technique to handle both the early and late stages of the DDBS design. To ensure that each operation flows into the next without any increase in complexity, fragmentation and allocation are done simultaneously. With this approach, the main goals are to minimize communication expenses, response time, and irrelevant data access. Most importantly, it has been observed that the proposed approach may effectively expand RDDBS performance by simultaneously fragmenting and assigning various relations. Through simulations and experiments on synthetic and real databases, we demonstrate the viability of our strategy and how it considerably lowers communication costs for typical access patterns at both the early and late stages of design.

Collaborative filtering (CF), one of the most widely employed methodologies for recommender systems, has drawn undeniable attention due to its effectiveness and simplicity. Nevertheless, a few papers have been published on the CF-based item-based model using similarity measures than the user-based model due to the model's complexity and the time required to build it. Additionally, the substantial shortcomings in the user-based measurements when the item-based model is taken into account motivated us to create stronger models in this work. Not to mention that the common trickiest challenge is dealing with the cold-start problem, in which users' history of item-buying behavior is missing (i.e., new users) or items for which activity is not provided (i.e., new items). Therefore, our novel five similarity measures, which have the potential to solve sparse data, are developed to alleviate the impact of this important problem. Most importantly, a thorough empirical analysis of how the item-based model affects the CF-based recommendation system’s performance has also been a critical part of this work, which presents a benchmarking study for thirty similarity metrics. The MAE, MSE, and accuracy metrics, together with fivefold cross-validation, are used to properly assess and examine the influence of all considered similarity measures using the Movie-lens 100 K and Film Trust datasets. The findings demonstrate how competitive the proposed similarity measures are in comparison to their alternatives. Surprisingly, some of the top "state-of-the-art" performers (such as SMD and NHSM) have been unable to fiercely compete with our proposed rivals when utilizing the item-based model.