Citations: Google scholar, Research gate

Past Research

Self-configuring Architectures

Modern systems are pressured to adapt in response to their constantly changing environment to remain useful. Traditionally, this adaptation has been handled at down times of the system. there is an increased demand to automate this process and achieve it whilst the system is running. Self-adaptive systems were introduced as a realization of continuously adapting systems. Self-adaptive systems are able to modify at runtime their behavior and/or structure in response to their perception of the environment, the system itself, and their requirements. The focus of this work is on realizing self-configuration, a key and essential property of self-adaptive systems. Self-configuration is the capability of reconfiguring automatically and dynamically in response to changes. This may include installing, integrating, removing and composing/decomposing system elements. We introduce the Dr-BIP framework, an extension of the BIP framework for modeling self-configuring systems that relies on a model-based and component & connector approach to prescribe systems.

A Dr-BIP system model is a runtime model which captures the running system at three different levels of abstraction namely behavior, configuration, and configuration variants. To capture the three levels of abstraction, we introduce motifs as primary structures to prescribe a self-configuring Dr-BIP system. A motif defines a set of components that evolve according to interaction and reconfiguration rules. A system is composed of multiple motifs that possibly share components and evolve together. Interaction rules dictate how components composing the system can interact and reconfiguration rules dictate how the system configuration can evolve over time. Finally, we show that the proposed framework is both minimal and expressive by modeling four different self-configuring systems. Last but not least, we propose a modeling language to codify the framework concepts and provision an interpreter implementation.

Credibility Analysis

Data generated on Twitter has become a rich source for various data mining tasks. Those data analysis tasks that are dependent on the tweet semantics, such as sentiment analysis, emotion mining, and rumor detection among others, suffer considerably if the tweet is not credible, not real, or spam. We perform an extensive analysis on credibility of Arabic content on Twitter. We also build a classification model (CAT) to automatically predict the credibility of a given Arabic tweet. Of particular originality is the inclusion of features extracted directly or indirectly from the author’s profile and timeline. To train and test CAT, we annotated for credibility a data set of 9, 000 Arabic tweets that are topic independent. CAT achieved consistent improvements in predicting the credibility of the tweets when compared to several baselines and when compared to the state-of-the-art approach with an improvement of 21% in weighted average F-measure. We also conducted experiments to highlight the importance of the userbased features as opposed to the contentbased features. We conclude our work with a feature reduction experiment that highlights the best indicative features of credibility.

Selected Publications

All publications