Reem Bahgat

Smartphone Energizer: Extending Smartphone's battery life with smart offloading

This paper presents “Smartphone Energizer”, a novel technique for context-aware computation offloading for smartphones. Previous techniques to the offloading problem were based on a narrow set of contextual information, which made the amount of energy saving varies unexpectedly based on the context in which the application is running. Smartphone Energizer uses the benefit of supervised learning with a rich set of contextual information such as application, device, network, and user characteristics to optimize both the energy consumption and execution time of Smartphones applications in a variety of contextual situations. In our evaluation, we show that Smartphone Energizer predicts both energy consumption and execution time in different contexts with error less than 9%, which in turn helped in taking the right offloading decision and saving energy by 40% to 56% and execution time by 43% to 58%.

Non-deterministic concurrent logic programming in Pandora

Introduction logic programming Pandora - the language constraint-based reasoning in Pandora for solving resource allocation problems Pandora for distributed discrete event simulation the Pandora deadlock handler PANDA a prototype PANDorA system the Pandora abstract machine.

The agent visualization system: a graphical and textual representation for multi-agent systems

As scientists from various domains increasingly resort to agent-based simulation for a more thorough understanding of real-world phenomena, the need for a simulation environment that facilitates rapid development of multi-agent systems is growing. Such a platform should provide means of visualizing the simulated scenario. In this paper we present the agent visualization system, the first system of its kind to specifically focus on catering to the visualization needs of agent-based simulation. The proposed system is a generic add-on that equips a simulation environment with a rich set of visualization facilities offering a variety of textual and graphical browsers that allow the modeler to detect trends and relationships in the simulation scenario. Some techniques from the field of information visualization were adapted and added to the system, while others were devised especially to be used in it. Regardless of their origin, all visualization techniques were thoroughly revised to make them generic enough to fit in our generic system. Agent visualization is more challenging than traditional information visualization in more than one respect. One of them is that the data to be visualized is not static; the simulation system is constantly producing data with every time step. Moreover, the sheer amount of data, together with its diversity, call for special adaptations to ensure that the system remains responsive and generic. To illustrate the various features of the proposed agent visualization system, we present a visualization of MicroTerra; a simulation scenario involving a group of beings trying to maximize their food intake.

An expert system for the best weight distribution on ferryboats

Abstract There are some problems that need expertise in order to get a satisfactory solution. Ferryboat carries goods, fresh water, diesel oil, luggage and storing rooms up to its permissible draft in order to maintain safety according to the international safety regulations. The best weight distribution on ferryboat needs human expertise to handle many variables, such as the amount of the bunker and fresh water that allow us to use more rooms for charging in order to maximize the profit. This sort of problems can be classified under Configuration Problem. In this paper, we address the development of a ferryboat expert systems (WDFB) using CommonKADS knowledge engineering methodology. We propose a reusable problem-solving approach, which is an enhancement of the structure-oriented approach, capable of solving the ferryboat configuration problem. The proposed model includes heuristics that make the search of suitable configuration more efficient, taking into consideration the transformation knowledge and the optimality criteria. The results of testing the system on a real-world data from National Navigation Company, Suez, Egypt, were satisfactory.

A NEW EMAIL RETRIEVAL RANKING APPROACH

Email Retrieval task has recently taken much attention to help the user retrieve the email(s) related to the submitted query. Up to our knowledge, existing email retrieval ranking approaches sort the retrieved emails based on some heuristic rules, which are either search clues or some predefined user criteria rooted in email fields. Unfortunately, the user usually does not know the effective rule that acquires best ranking related to his query. This paper presents a new email retrieval ranking approach to tackle this problem. It ranks the retrieved emails based on a scoring function that depends on crucial email fields, namely subject, content, and sender. The paper also proposes an architecture to allow every user in a network/group of users to be able, if permissible, to know the most important network senders who are interested in his submitted query words. The experimental evaluation on Enron corpus prove that our approach outperforms known email retrieval ranking approaches..

Jini versus CORBA for distributed systems

This paper presents performance comparisons between the CORBA and Jini middleware. The goal of this work is to investigate the behavior of each middleware and discover classes of applications that may perform better under each one of them.

The integration among disambiguation lexical resources for more effective phrase-level contextual polarity recognition

Phrase-level polarity disambiguation recently became attractive. Nowadays, most corners of the sentiment analysis research have been investigated while the core and hard parts are not yet intensively explored, among which polarity disambiguation is one. In this research, we propose the integration between two SentiWordNet-based resources on phrase-level polarity disambiguation: (1) its score for the best word sense identified by a word sense disambiguation algorithm and (2) its examples having the non-zero scores of the prior polarities for the related subjectivity lexicon words. The integration of subjectivity lexicon word prior polarities together with Senti-WordNet scores and examples of word contexts were presented as classifier features to reduce the classification disambiguation. Our empirical evaluation is twofold. First, an experimental analysis was intrinsically conducted using various lexical resource settings coupled with feature selection algorithms to improve the classifier contextual subjective and sentiment polarity recognition. Second, the integration was extrinsically applied to improve the opinion question answering ranking task. These evaluations prove the superiority of the proposed integration in comparison with the state-of-the-art seminal work and baselines.

Improving Detection Accuracy in Group Testing-Based Identification of Misbehaving Data Sources

Internet of Things (IoT) is an emerging field of research, in which heterogeneous objects effectively communicate with each other over the Internet and each object has a unique address. The fact that objects can access each other makes security a major issue. One of the serious security threats is Denial-of-Service (DoS) attack. DoS attacks overwhelm their victims with fake requests to consume their resources so that they become incapable of serving their legitimate clients. One category of proposed DoS defense techniques uses the Group Testing theory for detecting the identities of DoS attackers among the client group quickly and with low state overhead. In this paper, a group-testing-based DoS defense algorithm, namely Live Baiting, is tested in action. A concrete implementation is used to verify the algorithms feasibility, effectiveness and weaknesses. The algorithm is modified by introducing dynamic threshold and tolerance degree to enhance its accuracy. Under a high volume of HTTP traffic, the modified algorithm exhibited a detection accuracy (in terms of F-measure) that is improved by up to 300% as compared to the original algorithm.

A UML Heavyweight Extension for MAS Modeling

Unified Modeling Language (UML) is the standard notation technique used for modeling object-oriented (OO) systems. Previous efforts exist to extend UML for representing multi-agent systems (MAS). Due to the UML widespread, the resulting extensions are therefore easy to use and to adopt by software engineers. Unfortunately, most of these attempts are based on applying stereotypes for the object-oriented entities. However, heavyweight extension is a more adequate choice to handle different capabilities of the MAS over the OO systems. In this work a heavyweight extension to UML metamodel is proposed for modeling MAS according to our unified MAS conceptual model. It is not limited to a specific application domain or specific agent architecture. A proposed MAS-UML tool is built based on the extended metamodel.

Using mozart for visualizing agent-based simulations

Scientists from various domains resort to agent-based simulation for a more thorough understanding of complex real-world systems. We developed the Agent Visualization System; a generic system that can be added to a simulation environment to enrich it with a variety of browsers allowing the modeler to gain insight into his simulation scenario. In this paper we discuss how the various features of the Oz language and the Mozart platform aided us in the development of our system. Of particular importance were dataflow variables, high-orderness, the support for distribution and concurrency, the flexibility offered by QTk which was crucial in generating browsers whose structure is only known at run-time, in addition to a miscellany of features that were conductive to our work. We also highlight some of the implementation difficulties we faced and explain the techniques we utilized in overcoming them.