Drexel Celebrates Philly Tech Week 2014 at Inaugural CCI Day

Last Friday, April 11, 2014, Drexel University faculty, students, professional staff and alumni gathered in Behrakis Grand Hall to celebrate CCI Day—the first-ever exclusive showcase of College of Computing & Informatics’ (CCI) outstanding research and academic achievements.

Through poster presentations and interactive demonstrations, the showcase exhibited the breadth of the College’s innovative computing and information research in areas such as mobile computing, game development, cybersecurity, software engineering, informatics, and more.

CCI Day was an official event for Philly Tech Week 2014—a week-long celebration of technology and innovation, organized by Technically Philly and presented by AT&T, from April 4–12, 2014.

As the College of Computing & Informatics shares its acronym “CCI” with the Roman numeral for 201, CCI Day is held annually on the 201st day of the academic year.

Attendees also competed in a round of “CCI Jeopardy” led by Dr. William Regli, professor & senior associate dean for CCI Research and Scholarly Activities, in which participants submitted answers to trivia questions on informatics and computing topics for a chance to win various giveaways.

Students in the Ph.D. in computer science and information studies programs presented their research posters to a panel of judges—including CCI faculty members Drs. Larry Alexander, Ko Nishino, Il-Yeol Song and Kristene Unsworth—to compete for CCI Day Ph.D. Poster Awards (for computer science or information studies), and the People’s Choice Award.

Doctoral students, Aylin Caliskan-Islam and Mengwen Liu, took home the CCI Day Ph.D. Poster Awards in computer science and information studies, respectively. Caliskan’s project titled “Doppelgänger Finder: Taking Stylometry To The Underground” explores stylometric methods’ performance when applied to the cybercrime underworld (see abstract). Liu’s project titled “Semantic Biomedical Relation Extraction from Scientific Literature” uses biomedical literature to discover likely relationships between two types of biomedical entities, gene and brain regions, which can be used to construct a heterogeneous biological semantic network that benefits applications such as gene finding for specific brain regions and gene interaction analysis (see abstract).

Computer science doctoral student Linge Bai won the People’s Choice Award for project titled “Predicting Spatial Self-Organization with Statistical Moments,” in which she developed a self-organizing shape formation system based on locally interacting agents whose behaviors are inspired by living cells (see abstract).


CCI Day Doctoral Award Winner Abstracts

INAUGURAL CCI DAY PH.D. POSTER AWARD- Computer Science:  Aylin Caliskan-Islam

Title:Doppelgänger Finder: Taking Stylometry To The Underground

Abstract: Stylometry is a method for identifying anonymous authors of anonymous texts by analyzing their writing style. While stylometric methods have produced impressive results in previous experiments, we wanted to explore their performance on a challenging dataset of particular interest to the security research community. Analysis of underground forums can provide key information about who controls a given bot network or sells a service, and the size and scope of the cybercrime underworld. Previous analyses have been accomplished primarily through analysis of limited structured metadata and painstaking manual analysis. However, the key challenge is to automate this process, since this labor intensive manual approach clearly does not scale. We consider two scenarios. The first involves text written by an unknown cybercriminal and a set of potential suspects. This is standard, supervised stylometry problem made more difficult by multilingual forums that mix l33t-speak conversations with data dumps. In the second scenario, you want to feed a forum into an analysis engine and have it output possible doppelgängers, or users with multiple accounts. While other researchers have explored this problem, we propose a method that produces good results on actual separate accounts, as opposed to data sets created by artificially splitting authors into multiple identities. For scenario 1, we achieve 77% to 84% accuracy on private messages. For scenario 2, we achieve 94% recall with 90% precision on blogs and 85.18% precision with 82.14% recall for underground forum users. We demonstrate the utility of our approach with a case study that includes applying our technique to the Carders forum and manual analysis to validate the results, enabling the discovery of previously undetected doppelgänger accounts."


INAUGURAL CCI DAY PHD POSTER AWARD- Information Studies: Mengwen Liu

Title: Semantic Biomedical Relation Extraction from Scientific Literature

Abstract: Scientific literature contains abundant knowledge about relationships among concepts or entities. Unfortunately, such kind of knowledge is expressed in natural language where different types of relationships are not explicitly categorized. In this study, we take biomedical literature from Elsevier Neuroscience corpus as an example to discover likely relationships between two types of biomedical entities, gene and brain region. We develop a novel relation extraction approach that integrates distant learning and open information extraction techniques. Unlike state-of-the-art models of relation extraction from biomedical literature which are based on supervised learning, our approach does not need manually-labeled examples of relations. We apply our approach to extract relationships between genes and brain regions from one million articles. The resultant relations can be used to construct a heterogeneous biological semantic network. This network can benefit many interesting applications such as gene finding for specific brain regions and gene interaction analysis.



Title:Predicting Spatial Self-Organization with Statistical Moments

Abstract: We have developed a self-organizing shape formation system based on locally interacting agents whose behaviors are inspired by living cells. Given a predefined macroscopic shape, genetic programming is used to find a finite field function that defines the agents’ interactions. By following the gradient of the cumulative field the agents form into a desired shape. It has been seen that the self-organization process may form two or more stable final configurations. In order to control the outcome of the shape formation process, it is first necessary to accurately predict the outcome of the dynamic simulation. This paper describes an approach to predicting the final configurations produced by our spatial self-organization system at an early stage in the process. The approach calculates statistical moments of the coordinates of the agents, and employs Support Vector Machines to predict the final shape of the agent swarm based on the moments and their time derivatives.

Related News

Contact Us

Have a question? We’re eager to talk with you.

Contact Us