Capstone Final Exam Schedule

 

Final Examination Schedule

PLEASE JOIN US AS THE FOLLOWING CANDIDATES PRESENT THEIR CULMINATING WORK.

Autumn 2016

 

Tuesday, November 22nd

Duc-Huy Do
Chair: Dr. Brent Lagesse
Candidate: Master of Science in Computer Science & Software Engineering
3:00 PM; DISC 464
Application of UW Agents for Secure Communication Systems

The increase of distributed computing has led to the development of many different paradigms for network communication between hosts. Specifically the concept of mobile computing allows migration of programs to be run on other machines; these “mobile agent” objects are implemented to be autonomous, which brings many benefits such as network bandwidth management or compile time reduction. However, a significant drawback of code mobility is its vulnerability to attacks; for instance, a malicious agent could potentially harm a trusted host. This project consists of building a framework to develop applications with secure code mobility; built upon UW Agent from the University of Washington, the security scheme implemented provides data authentication and confidentiality for every communication done through the agents.   

Thursday, December 1st

Chris Kahne
Chair: Dr. Geethapriya Thamilarasu
Candidate: Master of Science in Cyber Security Engineering
1:00 PM; DISC 464
Design and Implementation of a Smart Home Intrusion Detection System

Smart home systems are increasingly used in various applications from home automation, security monitoring to remote control of devices connected across a home environment. While the ubiquitous computing has popularized these devices, security is increasingly a serious concern in these systems. Malicious entities can easily manipulate these devices to obtain detailed information about our physical home, such as detecting when a home is unoccupied and then unlocking the door for thieves.  In enterprise computing systems, Intrusion Detection Systems (IDS) have been used to identify when malicious behavior is occurring, but these IDS have not been introduced into smart homes due to the computer processing and power limitations of typical smart home devices.

This project proposes a dedicated IDS system to identify and detect common attacks in a smart home environment. In particular, a specification based approach is used to monitor behavioral patterns of the network to identify malicious behavior while minimizing the power and computing burden on the existing network.  The proposed IDS provides the foundation for individual or pre-packaged smart home intrusion detection solutions by providing an effective, maintainable, scalable, and extensible solution.        

Friday, December 2nd

Aytul Arisoy
Chair: Dr. David Socha
Candidate: Master of Science in Computer Science & Software Engineering
1:00 PM; DISC 464
Exploratory Wayfinding in Wide-Field Ethnography

This thesis focuses on what wayfinding (WF) means in relation to Wide-Field Ethnography (WFE), and how WF activities can be performed in WFE datasets.  The primary research question is: “What is Exploratory Wayfinding in WFE and how can a system help enabling WFE researchers do Exploratory Wayfinding in WFE?” WFE currently presents a gap, since WFE datasets are large and complex for researchers to easily understand, navigate, browse, filter, annotate, and analyze without framework and tool support. This thesis studies how tools or processes can help WFE researchers find their way through large, complex, unorganized, and continuously changing datasets, so that they can find the portions to analyze more easily. Using empirical data collected from analysis sessions and experiments within the context of qualitative research and exploratory data analysis, this research identifies a set of WFE WF hassles and proposes approaches by which these hassles could be mitigated to simplify and improve WFE WF. In particular, this thesis, defines a new concept called Exploratory Wayfinding (EW) in WFE, which aims to use tools and/or automation to augment human capabilities, and help researchers navigate and explore WFE datasets. In order to evaluate the WF concepts in WFE, we built a tool for EW in WFE, and then used a sequence of human centered design loops to help us evaluate the efficacy of our approach. Finally, this research argues that efficient wayfinding in WFE datasets is done via a combination of automated tools and a researcher’s innate senses.                 

Tuesday, December 6th

Wei Wang
Chair: Dr. David Socha
Candidate: Master of Science in Computer Science & Software Engineering
12:00 PM; DISC 464
Building cloud based software systems to support collaborative analysis of large, multi-stream, multi-modal datasets of physical-cyber-social systems

Traditional computer assisted qualitative data analysis (CAQDAS) tools have been lacking to take advantage of recent advancement of cloud computing services. Our challenges in effectively managing and analyzing the datasets of multiple-stream, multi-modal physical-cyber-social systems have motivated us to develop a modern set of tools to help researchers collaboratively analyze them. The primary goals are to better address the need for management and tracking provenance of datasets that are collected using multiple types of sensors to a variety of different data mediums, such as video, audio, images and text. Cloud-based technologies enable us to design toward a seamless collaboration experience across multiple geolocations and devices, in addition to taking advantage of state of art data security and integrity features provided by the cloud. The new generation CAQDAS tools we named WFE tools that we propose in this report, work in cooperation with Transana, and leverage the modern web technology and cloud infrastructure and services. Our system design, development challenges and current state and results of the built prototype tools are hereby presented.

 

Questions? Please email Megan Jewell, CSS Graduate Advisor at mjewell@uw.edu