Amr Badr

A proof of convergence for Ant algorithms

A proof of convergence for Ant algorithms is developed. Ant algorithms were modeled as branching random processes: the branching random walk and branching Wiener process to derive rates of birth and death of ant paths. Substitution is then carried out in birth-death processes which proves that a stable distribution is surely reached. This indicates that Ant algorithms converge with probability one. This analogy models Ant algorithms complexity parameters such as the number of cycles, the degrees of freedom of problem and the number of ants.

A binary clonal flower pollination algorithm for feature selection

This paper proposes a new hybrid algorithm that combines the Clonal Selection Algorithm with the Flower Pollination Algorithm.The proposed algorithm is applied to solve the feature selection problem.The accuracy of the Optimum Path Forest (OPF) classifier used as the objective function.Experiments have been implemented on three public benchmark datasets.Results demonstrated that the proposed hybrid algorithm is promising comparing with some well-known algorithms. Feature selection problem has been detected essentially in the last years. It is a step that is considered the prerequisite of the classification step. For the feature selection problem, the goal is to find out the most important subset of features that represent the original features in a certain domain. The selected features are used in optimization of a certain fitness function, so the feature selection problem can be seen as an optimization problem. This paper presents a new hybrid algorithm that combines Clonal Selection Algorithm (CSA) with Flower Pollination Algorithm (FPA) to compose Binary Clonal Flower Pollination Algorithm (BCFA) to solve the feature selection problem. The accuracy of the Optimum-Path Forest (OPF) classifier is used as an objective function. The experiments were implemented on three public datasets and demonstrated that the proposed hybrid algorithm achieved remarkable results in comparison with other well-known algorithms such as Binary Cuckoo Search Algorithm (BCSA), Binary Bat Algorithm (BBA), Binary Differential Evolution Algorithm (BDEA) and Binary Flower Pollination Algorithm (BFPA).

Image classification and retrieval using optimized Pulse-Coupled Neural Network

We propose an image classification and retrieval technique using PCNN and K-NN.We optimized the PCNN parameters using genetic algorithm.We implemented a prototype to validate our proposed technique.The results are represented and measured with precision, recall and accuracy.The proposed technique proved its efficiency in classifying and retrieving images with comparison to other techniques. Content-Based Image Retrieval (CBIR) has become a powerful tool that is used in many image applications and search engines. Thus, many techniques and approaches for CBIR were developed in literature. The CBIR approach works on the visual features of the image rather than a descriptive text. Therefore, it provides more effective and efficient retrieval. On the other hand, PCNN has proved its efficiency as an image processing tool for various tasks such as image segmentation and recognition, feature extraction, edge and object detection. This article introduces a technique for content-based image classification and retrieval using PCNN. The proposed technique uses an optimized Pulse-Coupled Neural Network (PCNN) to extract the visual features of the image in a form of a numeric vector called image signature. An optimization mechanism was applied to the PCNN parameters in order to improve the signature quality. Thus improving the classification and retrieval results. Additionally, it employs the K-Nearest Neighbor (K-NN) algorithm for classification and matching. By applying classification before retrieval, the number of images in the search space is optimized to include one category instead of multiple categories. Moreover, we developed a CBIR prototype to validate our technique. The results show that our technique can retrieve and classify images efficiently. Furthermore, we evaluated our prototype against one of the widely used techniques and it was proven that the proposed technique can enhance the search results and improve the accuracy by 3.5%.

One Rank Cuckoo Search Algorithm with Application to Algorithmic Trading Systems Optimization

uckoo search is a nature-inspired metaheuristic algorithm, based on the brood parasitism of some cuckoo species, along with Levy flights random walks. In this paper, a modified version is proposed, where the new solutions generated from the exploration and exploitation phases are combined, evaluated and ranked together, rather than separately in the original algorithm, in addition to imposing a bound by best solutions mechanism to help improve convergence rate and performance. The proposed algorithm was tested on a set of ten standard benchmark functions, and applied to a real-world problem of algorithmic trading systems optimization in the financial markets. Experimental analysis demonstrated improved performance in almost all benchmark functions and the problem under study.

Using machine learning techniques for evaluating tomato ripeness

Egypt occupied the fifth place in both income and weight of tomato production.We proposed an automated multi-class classification approach for tomato ripeness stages.Performance of classification algorithms depends on statistics of the experimented dataset.Training and testing datasets have been generated via employing the 10-fold cross validation.Using OAO multi-class SVMs with linear kernel function outperformed other algorithms. Tomato quality is one of the most important factors that helps ensuring a consistent marketing of tomato fruit. As ripeness is the main indicator for tomato quality from customers perspective, the determination of tomato ripeness stages is a basic industrial concern regarding tomato production in order to get high quality product. Automatic ripeness evaluation of tomato is an essential research topic as it may prove benefits in ensuring optimum yield of high quality product, this will increase the income because tomato is one of the most important crops in the world. This article presents an automated multi-class classification approach for tomato ripeness measurement and evaluation via investigating and classifying the different maturity/ripeness stages. The proposed approach uses color features for classifying tomato ripeness stages. The approach proposed in this article uses Principal Components Analysis (PCA) in addition to Support Vector Machines (SVMs) and Linear Discriminant Analysis (LDA) algorithms for feature extraction and classification, respectively. Experiments have been conducted on a dataset of total 250 images that has been used for both training and testing datasets with 10-fold cross validation. Experimental results showed that the proposed classification approach has obtained ripeness classification accuracy of 90.80%, using one-against-one (OAO) multi-class SVMs algorithm with linear kernel function, ripeness classification accuracy of 84.80% using one-against-all (OAA) multi-class SVMs algorithm with linear kernel function, and ripeness classification accuracy of 84% using LDA algorithm.

CLUSTERING TWEETS USING CELLULAR GENETIC ALGORITHM

As the popularity of Twitter continues to increase rapidly, it is extremely necessary to analyze the h uge amount of data that Twitter users generate. A popul ar method of tweet analysis is clustering. Because most tweets are textual, this study focuses on clusterin g tweets based on their textual content similarity. This study presents tweet clustering using cellular gene tic algorithm cGA. The results obtained by cGA are compared with those obtained by generational geneti c algorithm in terms of average fitness, average ti me required for execution and number of generations. E xperimental results are tested with two sets: One o f 1000 tweets and the second formed of 5000 tweets. T he results show a nearly equal performance for both algorithms in terms of the average fitness of the s olution. On the other hand, cGA shows a much faster performance than generational. These results demonstrate that cellular genetic algorithm outperforms generational genetic algorithm in tweet clustering.

HYBRID ANT-BASED CLUSTERING ALGORITHM WITH CLUSTER ANALYSIS TECHNIQUES

Cluster analysis is a data mining technology designed to derive a good understanding of data to solve clustering problems by extracting useful information from a large volume of mixed data elements. Recently, researchers have aimed to derive clustering algorithms from nature’s swarm behaviors. Ant-based clustering is an approach inspired by the natural clustering and sorting behavior of ant colonies. In this research, a hybrid ant-based clustering method is presented with new modifications to the original ant colony clustering model (ACC) to enhance the operations of ants, picking up and dropping off data items. Ants’ decisions are supported by operating two cluster analysis methods: Agglomerative Hierarchical Clustering (AHC) and density-based clustering. The proximity function and refinement process approaches are inspired by previous clustering methods, in addition to an adaptive threshold method. The results obtained show that the hybrid ant-based clustering algorithm attains better results than the ant-based clustering Handl model ATTA-C, k-means and AHC over some real and artificial datasets and the method requires less initial information about class numbers and dataset size.

A Service Oriented Architecture for Biomedical Question Answering System

Medical practitioners consider online medical resources as one of the important sources to answer their biomedical questions. Finding the answer to a certain question among large amount of medical documents - available on the Web - is very time consuming. In this paper we present a biomedical question answering system that is based on service oriented architecture and Web services. Service oriented architecture facilitates this system development by providing modular design, loosely coupled application integration, asset reuse, system agility, and interoperability.

DBCLUM: Density-based Clustering and Merging Algorithm

Clustering is a primary method for DB mining. The clustering process becomes very challenge when the data is different densities, different sizes, different shapes, or has noise and outlier. Many existing algorithms are designed to find clusters. But, these algorithms lack to discover clusters of different shapes, densities and sizes. This paper presents a new algorithm called DBCLUM which is an extension of DBSCAN to discover clusters based on density. DBSCAN can discover clusters with arbitrary shapes. But, fail to discover different-density clusters or adjacent clusters. DBCLUM is developed to overcome these problems. DBCLUM discovers clusters individually then merges them if they are density similar and joined. By this concept, DBCLUM can discover different-densities clusters and adjacent clusters. Experiments revealed that DBCLUM is able to discover adjacent clusters and different-densities clusters and DBCLUM is faster than DBSCAN with speed up ranges from 11% to 52%.

Prediction of O-glycosylation Sites Using Random Forest and GA-Tuned PSO Technique

O-glycosylation is one of the main types of the mammalian protein glycosylation; it occurs on the particular site of serine (S) or threonine (T). Several O-glycosylation site predictors have been developed. However, a need to get even better prediction tools remains. One challenge in training the classifiers is that the available datasets are highly imbalanced, which makes the classification accuracy for the minority class to become unsatisfactory. In our previous work, we have proposed a new classification approach, which is based on particle swarm optimization (PSO) and random forest (RF); this approach has considered the imbalanced dataset problem. The PSO parameters setting in the training process impacts the classification accuracy. Thus, in this paper, we perform parameters optimization for the PSO algorithm, based on genetic algorithm, in order to increase the classification accuracy. Our proposed genetic algorithm-based approach has shown better performance in terms of area under the receiver operating characteristic curve against existing predictors. In addition, we implemented a glycosylation predictor tool based on that approach, and we demonstrated that this tool could successfully identify candidate glycosylation sites in case study protein.