Showing papers by "Hossam Faris published in 2014"

Artificial Neural Networks for Surface Ozone Prediction: Models and Analysis

Abstract: Ozone is one of the most important constituents of the Earth's atmosphere. Ozone is vital because it maintains the thermal structure of the atmosphere. However, exposure to high concentrations of Ozone can cause serious problems to human health, vegetation, and damage to surfaces. The complexity of the relationship between the main attributes that severely affect surface ozone levels have made the problem of predicting its concentration very challenging. Innovative mathematical modeling techniques are urgently needed to get a better understanding of the dynamics of these attributes. In this paper, prediction of the surface ozone layer problem is investigated. A comparison between two types of artificial neural networks (ANN) (multilayer perceptron trained with backpropagation and radial basis functions (RBF) networks) for short prediction of surface ozone is conclusively demonstrated. Two models that predict the expected values of the surface ozone based on three variables (i.e. nitrogen-di-oxide, temperature, and relative humidity) are developed and compared.

Data Security Issues and Challenges in Cloud Computing: A Conceptual Analysis and Review

Abstract: Cloud computing is a set of Information Technology services offered to users over the web on a rented base. Such services enable the organizations to scale-up or scale-down their in-house foundations. Generally, cloud services are provided by a third-party supplier who possesses the arrangement. Cloud computing has many advantages such as flexibility, efficiency, scalability, integration, and capital reduction. Moreover, it provides an advanced virtual space for organizations to deploy their applications or run their operations. With disregard to the possible benefits of cloud computing services, the organizations are reluctant to invest in cloud computing mainly due to security concerns. Security is one of the main challenges that hinder the growth of cloud computing. At the same time, service providers strive to reduce the risks over the clouds and increase their reliability in order to build mutual trust between them and the cloud customers. Various security issues and challenges are discussed in this research, and possible opportunities are stated.

A Genetic Programming Based Framework for Churn Prediction in Telecommunication Industry

Abstract: Customer defection is critically important since it leads to serious business loss. Therefore, investigating methods to identify defecting customers (i.e. churners) has become a priority for telecommunication operators. In this paper, a churn prediction framework is proposed aiming at enhancing the ability to forecast customer churn. The framework combine two heuristic approaches: Self Organizing Maps (SOM) and Genetic Programming (GP). At first, SOM is used to cluster the customers in the dataset, and then remove outliers representing abnormal customer behaviors. After that, GP is used to build an enhanced classification tree. The dataset used for this study contains anonymized real customer information provided by a major local telecom operator in Jordan. Our work shows that using the proposed method surpasses various state-of-the-art classification methods for this particular dataset.

Improving production quality of a hot-rolling industrial process via genetic programming model

Abstract: Satisfying the customers' need for manufacturing plants and the demand for high-quality products becomes more challenging nowadays. Manufacturers need to retain advanced attributes of their products by applying high-quality automation process. In this paper, a genetic programming GP approach is applied in order to develop three mathematical models for the force, torque and slab temperature in the hot-rolling industrial process. A frequency-based analysis using GP is performed to provide an insight into the process significant factors. The performance of the GP developed models is evaluated with respect to the known soft computing models explored in the literature. Experimental data were collected from the Eregli Iron and Steel Factory in Turkey and used to test the performance of the GP models. Genetic programming shows better performance modelling capabilities compared with models-based artificial neural networks and fuzzy logic.

Neighborhood Cleaning Rules and Particle Swarm Optimization for Predicting Customer Churn Behavior in Telecom Industry

Abstract: Churn prediction is an important task for Customer Relationship Management (CRM) in telecommunication companies. Accurate churn prediction helps CRM in planning effective strategies to retain their valuable customers. However, churn prediction is a complex and challenging task. In this paper, a hybrid churn prediction model is proposed based on combining two approaches; Neighborhood Cleaning Rules (NCL) and Particle Swarm Optimization (PSO). NCL is applied in the preprocessing stage for handling the imbalanced churn data; and eliminating outliers and unrepresentative data. In the next stage, a Constricted PSO is applied for developing the final prediction model. The developed model is evaluated and compared with a baseline PSO model. The proposed hybrid model is compared also with Artificial Neural Networks (ANN) and Decision trees (DT) models which are traditional and common approaches used in the literature for churn prediction. The experimental results show that the proposed hybrid model outperforms the baseline PSO model, ANN and DT in terms of accuracy and actual churn rate.

A Survey of Concepts Location Enhancement for Program Comprehension and Maintenance

Abstract: When correcting a fault, adding a new concept or feature, or adapting a system to conform to a new platform, software engineers must first find the relevant parts of the code that correspond to a particular change This is termed as concept or feature location process Several techniques have been introduced which automate some or all of the process of concept location Those techniques rely heavily on code comprehension as it is considered a prerequisite when attempting to maintain any software system It provides a comprehensive overview of large body work which is beneficial to researchers and practitioners This paper presents an overview of code comprehension categorization and consequence A systematic literature survey of concept location enhancement techniques is also presented Moreover, the paper presents an overview of the role of concept location in program comprehension and maintenance and discusses information retrieval techniques to advance concept location

Visualization analysis of feed forward neural network input contribution

Abstract: The complexity of domain problem can slow or even hinder the learning process of neural networks. It is rather difficult to overcome such an obstacle because neural networks, as cited today in the literature, lack the interpretability of their internal structures. In this paper, we present a visualization approach capable of enhancing the understanding of neural networks. Our approach visualizes input and weight contributions, sensitivity analysis, and provides guidance in pruning less influential features and consequently reducing the complexity of domain problem while maintaining acceptable error rates. We conduct experiments on various datasets to show the effectiveness of our approach. Key words: Neural network, visualization, input contribution, sensitivity analysis

On Symbolic Regression for Optimizing Thermostable Lipase Production

Abstract: Theromostable lipases have wide range of biotechnological applications in the industry. Therefore, there is always high interest in investigating their features and operating conditions. However, Lipase production is a challenging and complex process due to its nature which is highly dependent on the conditions of the process such as temperature, initial pH, incubation period, time, inoculum size and agitation rate. Efficient optimization of the process is a common goal in order to improve the productivity and reduce the costs. In this paper, we apply a Symbolic Regression Genetic Programming (GP) approach in order to develop a mathematical model which can predict the lipase activities in submerged fermentation (SmF) system. The developed GP model is compared with a neural network model proposed in the literature. The reported evaluation results show superiority of GP in modeling and optimizing the process.

Shaping an experts locator system: recommending the right expert

Abstract: Experiences and knowledge inside any organization is a high valued resource. In order to utilize this resource, organizations should facilitate the knowledge sharing process between experts and others. Experts-locator systems recommend experts within the organization. However, current systems does not take into account the responsiveness of experts in providing support when requested. The proposed system ranks and recommends experts based on factors such as experience duration, experience level, number of projects, and more importantly, number of sent emails as a responsiveness indicator. Delphi technique was followed to identify and weight the factors. The prototype system has been experimented and results indicate that ranking formula is useful to recommend more responsive experts.