Research Title: An efficient evolutionary algorithm with a nearest neighbor search technique for clustering analysis
Author: Raneem Nadim Qaddoura, Published Year: 2020
Journal of Ambient Intelligence and Humanized Computing,
Faculty: Information Technology

Abstract: Evolutionary algorithms have shown their powerful capabilities in different machine learning problems including clustering which is a growing area of research nowadays. In this paper, we propose an efficient clustering technique based on the evolution behavior of genetic algorithm and an advanced variant of nearest neighbor search technique based on assignment and election mechanisms. The goal of the proposed algorithm is to improve the quality of clustering results by finding a solution that maximizes the separation between different clusters and maximizes the cohesion between data points in the same cluster. Our proposed algorithm which we refer to as “EvoNP” is tested with 15 well-known data sets using 5 well-known external evaluation measures and is compared with 7 well-regarded clustering algorithms . The experiments are conducted in two phases: evaluation of the best fitness function for the algorithm and evaluation of the algorithm against other clustering algorithms. The results show that the proposed algorithm works well with silhouette coefficient fitness function and outperforms the other algorithms for the majority of the data sets. The source code of EvoNP is available at http://evo-ml.com/evonp/.

Keywords: Cluster analysis, Clustering, Nearest neighbor search, Evolutionary algorithms, Optimization algorithms, Nature-inspired algorithms

Research Title: EvoCluster: An Open-Source Nature-Inspired Optimization Clustering Framework in Python
Author: Raneem Nadim Qaddoura, Published Year: 2020
International Conference on the Applications of Evolutionary Computation (Part of EvoStar), Spain
Faculty: Information Technology

Abstract: EvoCluster is an open source and cross-platform framework implemented in Python which includes the most well-known and recent nature-inspired metaheuristic optimizers that are customized to perform partitional clustering tasks. The goal of this framework is to provide a user-friendly and customizable implementation of the metaheuristic based clustering algorithms which can be utilized by experienced and non-experienced users for different applications. The framework can also be used by researchers who can benefit from the implementation of the metaheuristic optimizers for their research studies. EvoCluster can be extended by designing other optimizers, including more objective functions, adding other evaluation measures, and using more data sets. The current implementation of the framework includes ten metaheristic optimizers, thirty datasets, five objective functions, and twelve evaluation measures. The source code of EvoCluster is publicly available at (http://evo-ml.com/2019/10/25/evocluster/).

Keywords: Clustering, Cluster analysis, Evolutionary computing, Framework, Python

Research Title: Dental radiography segmentation using expectation-maximization clustering and grasshopper optimizer
Author: Raneem Nadim Qaddoura, Published Year: 2020
Multimedia Tools and Applications,
Faculty: Information Technology

Abstract: Image segmentation is a popular technique that is used for extracting information from images, which has also gained a lot of interest lately due to its importance in different scientific fields such as the medical field. This paper proposes a novel image segmentation technique using Expectation-Maximization (EM) clustering algorithm and Grasshopper Optimizer Algorithm (GOA). The proposed technique and the concept of image segmentation are effectively applied on dental radiography datasets that are collected from 120 patients with an age between 6 to 60 years old. To validate the proposed technique, a comparison in terms of purity and entropy measures is conducted against K-means, X-means, EM, and Farthest First algorithms. Based on our experimental results, the proposed technique using EM and GOA achieved the best results compared to other algorithms for all three datasets in terms of entropy and purity. The best results were obtained using the second dataset, which achieved purity value of 0.7126 and entropy value of 0.3083. Further, the proposed technique also outperforms U-net and Random Forest algorithms for the selected datasets.

Keywords: Image segmentation, Expectation-Maximization algorithm, Grasshopper optimization algorithm, Dental radiography, Anatomical segmentation and classification

Research Title: Intelligent detection of hate speech in Arabic social network: A machine learning approach
Author: Raneem Nadim Qaddoura, Published Year: 2020
Journal of Information Science,
Faculty: Information Technology

Abstract: Nowadays, cyber hate speech is increasingly growing, which forms a serious problem worldwide by threatening the cohesion of civil societies. Hate speech relates to using expressions or phrases that are violent, offensive or insulting for a person or a minority of people. In particular, in the Arab region, the number of Arab social media users is growing rapidly, which is accompanied with high increasing rate of cyber hate speech. This drew our attention to aspire healthy online environments that are free of hatred and discrimination. Therefore, this article aims to detect cyber hate speech based on Arabic context over Twitter platform, by applying Natural Language Processing (NLP) techniques, and machine learning methods. The article considers a set of tweets related to racism, journalism, sports orientation, terrorism and Islam. Several types of features and emotions are extracted and arranged in 15 different combinations of data. The processed dataset is experimented using Support Vector Machine (SVM), Naive Bayes (NB), Decision Tree (DT) and Random Forest (RF), in which RF with the feature set of Term Frequency-Inverse Document Frequency (TF-IDF) and profile-related features achieves the best results. Furthermore, a feature importance analysis is conducted based on RF classifier in order to quantify the predictive ability of features in regard to the hate class.

Keywords: Hate speech, machine learning, text vectorization, Twitter

Research Title: Predicting Students Performance in Online Courses using Classification Techniques
Author: Iman Saleh Naji, Published Year: 2020
The International Conference on Intelligent Data Science Technologies and Applications (IDSTA2020), Online Presentations
Faculty: Information Technology

Abstract: Data Mining techniques have become widely used for several types of tasks, such as data classification. Data classification aims to classify data into set of classes based on trained model called classifier. Such techniques are used to extract important information needed to make decisions in different scopes, one of them is Education. This paper, applies several classification techniques on an academic dataset for online courses, to find the best classifier to predict the student performance, using certain features that might affect this performance. Classifiers used are Decision Tree, Artificial Neural Network, Support Vector Machine, and K-Nearest Neighbors. Experiments used real data and evaluated the models based on four performance measures: accuracy, precision, recall, and F-measure. Results show that Decision Tree and Artificial Neural Network achieved best results among other classifiers.

Keywords: Classification, Decision Tree (DT), Artificial Neural Network (ANN), Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Students Performance

Research Title: Privacy Preserving of Shared Data in Deep Learning
Author: Iman Saleh Naji, Published Year: 2019
2019 International Arab Conference on Information Technology (ACIT), Al Ain, United Arab Emirates, United Arab Emirates
Faculty: Information Technology

Abstract: In the age of Big Data the need of developing machine learning algorithms has increased. Such algorithms are used to extract valuable information needed for different types of sectors; health, education, financial ...etc. In many cases data has to be shared among several parties to guarantee better accuracy of the results of such algorithms. In these cases privacy of data would be questionable. In this paper a survey has been conducted on research that focused on Privacy Preserving techniques when applying deep learning algorithms on distributed or shared data.

Keywords: Deep Learning , Multi-Party , Encryption , Data-Preserving

Research Title: PSMD12 haploinsufficiency in a neurodevelopmental disorder with autistic features
Author: Raida W. Khalil, Published Year: 2017
The Second Human Genetic Conference, Mendelian Diseases & the Next Generation Sequencing” , , University of Philadelphia- Jordan
Faculty: Science

Abstract: PSMD12 haploinsufficiency in a neurodevelopmental disorder with autistic features

Keywords: PSMD12 autistic

Research Title: Immunological Changes in SECONDARY SJOGRAN SYNDROME(sSs) (comparative study)
Author: Raida W. Khalil, Published Year: 2014
17th International Congress on Oral Pathology and Medicine,, Istanbul, Turkey
Faculty: Science

Abstract: Immunological Changes in SECONDARY SJOGRAN SYNDROME(sSs) (comparative study)

Keywords: Immunological SECONDARY SJOGRAN SYNDROME(sSs)

Research Title: The association between Toll-Like Receptor 4(TLR4) gene polymorphism and Jordanian type 2
Author: Raida W. Khalil, Published Year: 2010
the International Journal of Arts & Sciences (IJAS) conference- , Bad Hofgastein’s Kungress Zentrum-Austria
Faculty: Science

Abstract: The polymorphisms Toll-like receptor and Type 2 diabetes among Jordanian Patients, Three SNPs studied

Keywords: TOll 4, Type 2 diabetes , Jordan

Research Title: Seroprevalence and zoonotic potential of Neospora species infection in Jordanian women with miscarriage
Author: Raida W. Khalil, Published Year: 2010
British Society for Parasitology Spring Meeting and Trypanosomiasis& LeishmaniasisSeminarCardiff, Cardiff University, Wales, UK
Faculty: Science

Abstract: Serological detection of Neospora spp. was carried out on 445 Jordanian women with miscarriage, using the indirect fluorescent antibody test and N. caninum antigen. The type of hospital, age, cat and dog contact, raw meat and wild plant consumptions, number of miscarriages and stillbirths, were tested as risk factors for the seroprevalence using univariable and multivariable logistic regression analyses. At cutoff titers of 1:200 and 1:400 the seroprevalences of Neospora were 25% and 6% for IgG and 12% and 7% for IgM, respectively. Neospora-IgG-seropositivity was associated with having a dog in the household or in the immediate environment with odds ratios of 2.6 and 4.1 respectively, and had stillbirth with an odds ratio of 0.1. Neospora-IgM-seropositivity was associated with women with miscarriage visiting a private hospital. The current results provide serological and epidemiological evidence for the emerging zoonotic potential of Neospora species infection to be categorized as a level one zoonosis.

Keywords: Neospora, miscarriage, Jordan