Sherif A. Elfayoumy

Anti-Spam Filtering Using Neural Networks and Baysian Classifiers

Electronic mail is inarguably the most widely used Internet technology today. With the massive amount of information and speed the Internet is able to handle, communication has been revolutionized with email and other online communication systems. However, some computer users have abused the technology used to drive these communications, by sending out thousands and thousands of spam emails with little or no purpose other than to increase traffic or decrease bandwidth. This paper evaluates the effectiveness of email classifiers based on the feedforward backpropagation neural network and Baysian classifiers. Results are evaluated using accuracy and sensitivity metrics. The results show that the feedforward backpropagation network algorithm classifier provides relatively high accuracy and sensitivity that makes it competitive to the best known classifiers. On the other hand, though Baysian classifiers are not as accurate they are very easy to construct and can easily adapt to changes in spam patterns.

Cross Dock Scheduling Using Genetic Algorithms

Cross docking is a distribution system in which the merchandise received at a warehouse or distribution center is not stocked but immediately prepared for onward shipment. In other words, cross docking is the transfer of inward deliveries from the point of reception directly to the point of delivery with limited or no interim storage. One way to reduce cost in a cross dock terminal is to park incoming and outgoing trucks so that the loads can be efficiently moved across the dock. This means that the distance from loading and unloading is the shortest possible distance. The problem with generating an efficient schedule for the door assignments is that for n incoming and m outgoing trucks there are n!*m! possible solutions. This paper describes a solution to the cross dock scheduling problem using genetic algorithms. To judge the efficiency and accuracy of this solution, four programs were developed to test every possible combination of for small problem sizes (four, five, six, and seven incoming trucks). The results of the efficiency and accuracy testing shows that using genetic algorithms to schedule cross dock trucking operations provides an accurate and timely solution.

Analyzing web service choreography specifications using colored Petri nets

Enacting cross-organizational business processes requires critical support for long-running and complex interactions involving multiple participants. The Web Services Choreography Description Language (WS-CDL) aims at facilitating just that, by providing means to describe correlated message exchanges among services geared towards achieving a business goal. While WSCDL specifications are machine-readable documents, they do not necessarily allow developers to determine--by direct inspection--whetheror not the patterns of message exchanges they stipulate do indeed describethe intended service behavior. In this research paper, we show how Colored Petri Nets (CPN) can be used to analyze WS-CDL documents in order to identify faults in the specification. We have developed a research prototype that assists in the creation of a CPN model from a given WS-CDL document. The CPN model generated is then analyzed using the formal verification environment and simulation capability provided by CPN-Tools. We provide a discussion on the analysis of an example WS-CDL document using this approach, as well as on the advantages and limitations of using CPN for analyzing WS-CDL specifications.

Web-based performance visualization of distributed discrete event simulation

This paper reports on an effort to adapt an existing distributed simulation visualization system to become Web accessible. The system was originally developed for performance visualization and experimentation with parameters affecting PDES systems using the time Warp protocols. This paper presents a model for converting legacy PDES systems to be Web accessible, and discusses the initial results from the conversion effort on this specific application. After finishing this work, we will be able to collect a wealth of data through the Web for future data mining, and to create an intelligent agent for performance tuning of time Warp applications.

Adaptive Cache Replacement:A Novel Approach

Cache replacement policies are developed to help insure optimal use of limited resources. Varieties of such algorithms exist with relatively few that dynamically adapt to traffic patterns. Algorithms that are tunable typically utilize off-line training mechanisms or trial-and-error to determine optimal characteristics. Utilizing multiple algorithms to establish an efficient replacement policy that dynamically adapts to changes in traffic load and access patterns is a novel option that is introduced in this article. A simulation of this approach utilizing two existing, simple, and effective policies; namely, LRU and LFU was studied to assess the potential of the adaptive policy. This policy is compared and contrasted to other cache replacement policies utilizing public traffic samples mentioned in the literature as well as a synthetic model created from existing samples. Simulation results suggest that the adaptive cache replacement policy is beneficial, primarily in smaller cache sizes.

From Returns to Tweets and Back: An Investigation of the Stocks in the Dow Jones Industrial Average

ABSTRACT A sizeable percentage of investors are using social media to obtain information about companies (Cogent Research [2008]). As a consequence, social media content about firms may have an impact on stock prices (Hachman [2011]). Various studies utilize social media content to forecast stock market-related factors such as returns, volatility, or trading volume. The objective of this article is to investigate whether a bidirectional intraday relationship between stock returns and volatility and tweets exists. The study analyzed 150,180 minute-by-minute stock price and tweet data for the 30 stocks in the Dow Jones Industrial Average over a random 13-day interval from June 2 to June 18, 2014 using a BEKK-MVGARCH methodology. Findings indicate that 87% of stock returns are influenced by lagged innovations of the tweets data, but there is little evidence to support that the direction is reciprocal, with only 7% of tweets being influenced by lagged innovations of the stock returns. Results further show that the lagged innovations from 40 percent of stock returns affect the current conditional volatility of the tweets, while 73 percent of tweets affect the current conditional volatility of stock returns. Moreover, there is strong evidence to suggest that the volatility originating from the returns to the tweets persists for 33 percent of stocks; the volatility originating from the tweets to the returns persists for 73 percent of stocks. Last, 53 percent of stocks exhibit both immediate and persistent impacts from returns to tweets, while 90 percent of stocks exhibit both immediate and persistent impacts from tweets to returns. These results may help traders achieve superior returns by buying and selling individual stocks or options. Also, asset and mutual fund managers may benefit by developing a social media strategy.

PQL: Protein Query Language

This paper introduces a Protein Query Language (PQL) for querying protein structures in an expressive yet concise manner. One of the objectives of the paper is to demonstrate how such a language would be beneficial to protein researchers to obtain in-depth protein data from a relational database without extensive SQL knowledge. The language features options such as limiting query results by key protein characteristics such as methyl donated hydrogen bond interactions, minimum and maximum phi and psi angles, repulsive forces, CH/Pi calculations, and other pertinent factors. A backend data model was designed to support storage and retrieval of protein primary and secondary sequences, atomic-level data, as well as calculations on said data. A relational DBMS is used as the persistent storage backend, with every effort made to ensure transparent portability to most relational database systems. In addition, front end applications can be developed to support retrieving, transforming, and preprocessing of information from the Research Collaboratory for Structural Bioinformatics (RCSB) into the backend data repository. The new language and associated architecture allow users to load additional protein files from RCSB into the database, issue standard queries to download pertinent data in user-friendly formats including CSV files, issue non-standard queries against secondary structures via the protein query language, and run error-detection routines against data in the database. Query results may include normalized or denormalized data, model and chain data, residue data, atom detail data, and primary as well as secondary structure data.

Automatic heart volume measurement from CMR images using ant colony optimization with iterative salient isolated thresholding

Summary Manual segmentation of CMR images is inefficient and inconsistent method for measuring ventricular volumes. In this study, a new technique (ACOISIT) is developed and implemented for measuring ventricular volumes. The technique is based on automatic delineation of bloodmyocardium border using ant colony optimization with salient isolated thresholding. The technique was implemented on datasets from eight volunteers and the results were compared to manual segmentation. ACOISIT showed good agreement with the gold standard and was faster and more consistent than manual segmentation. Background