Capstone Final Exam Schedule

 

Final Examination Schedule

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

Spring 2016

 

Monday, May 23rd

Priyadarshini Ragupathy
Chair: Dr. Hazeline Asuncion
Candidate: Master of Science in Computer Science & Software Engineering
11:00 AM; DISC 464
Stock Analyzer

The goal of this project is to build a stock tracking application that integrates social media sentiments to help investors make informed investment decisions. Stock Analyzer is an iOS application where users can create and track a watch list of stocks. It shows detailed real-time quotes and historic data for any stock. A unique feature of this application is the ability to view social media sentiment for the stock calculated using data aggregated from Twitter. All the data is surfaced on the application with rich visualization using appropriate charts and animations. The application is backed by services hosted in Microsoft Azure, providing availability and reliability guarantees. The services fetch market and Twitter data from external providers and use a NoSQL database back-end to store sentiment data and cache quotes/charts to provide low latency and responsiveness even when the external data providers are non-responsive.

Tuesday, May 24th

Hsin C. Chen
Chair: Dr. Kelvin Sung
Candidate: Master of Science in Computer Science & Software Engineering
10:30 AM; DISC 464
Assessing Student Engagement via IDE Instrumentation

Introductory programming classes are challenging and have led to low retention rate in the Computer Science majors. Game-Themed Computer Science (GTCS) project aims at integrating videogame creation in these courses as programming assignments. Students should be more engaged in the learning process if their assignments are entertaining. This project is designed to set up a proof-of-concept system to track programming sessions of students in introductory programming courses, specifically CSS161. Using post-analysis on the tracking data would provide insights into the pattern of students working with GTCS assignments, and potentials of in-depth understanding of the effectiveness of teaching with the GTCS curriculum. This project tracks patterns in student behavior while developing solutions to GTCS assignments, including: the number of times an IDE has been opened, total amount of time spent working with an IDE, the most common compiler errors, etc. In this way, the project explores approaches to gather relevant information, with results serving as a pathway for future investigation.

Wednesday, May 25th

Rashmi Ragupal
Chair: Dr. William Erdly
Candidate: Master of Science in Computer Science & Software Engineering
11:00 AM; DISC 464
Tools Development for Children's Vision Assessment, Data Sharing and Web Presence: An Integrative Approach

My work extends across three important areas in support of the children’s vision assessment and therapy project at UWB.  These areas include 1) an application used to collect field data that encompasses six tests to assess visuo-cognitive-motor skills in children (as outlined in “Red Flags for Elementary Teachers”); 2) design an initial prototype of the Educating Young Eyes (EYE) website that hosts various vision screening tools; and 3) redesign the 2016 EYE Conference website that is scheduled for November 5, 6 at our campus. These projects involve a wide range of activities ranging from architecture design, coding (Android, Java, HTML5/CSS, SQLite), and various software engineering principles and activities.  More specifically, my goals are to enable seamless and cost-effective data-collection of various vision screening activities using an Android-based tablet application, provide easy and efficient data storage/retrieval ability, and support efficient and effective data transfer/reporting in remote areas with limited internet connectivity.

Thursday, May 26th

Larisa Kocsis
Chair: Dr. Kelvin Sung
Candidate: Master of Science in Computer Science & Software Engineering
9:00 AM; DISC 464
Exploratory Data Visualization for College Affordability

The complexity of financial aid estimation and application process is a leading reason for which low-income high school students in the United States fail to continue their education. Data visualization provides a simplified and intuitive solution to making college affordability accessible to the public. This project introduces a 3-way approach to improving the financial estimation process. First, we gather and clean the data. Next, we generate a multi-view exploratory visualization system. Finally, we provide refinement criteria for the estimate accuracy. This approach addresses areas of opportunity identified in existing systems and builds on the ongoing simplification efforts of financial aid application.

David Drobesh
Chair: Dr. Hazeline Asuncion
Candidate: Master of Science in Computer Science & Software Engineering
1:00 PM; DISC 464
FACTS-IDE: Code Change Traceability Semantic Models integrated within Eclipse

Software maintenance expenses and code change costs are natural challenges that arise from successful software evolution, yet current models and techniques to manage these realities are overwhelming or insufficient. To address these challenges, we propose an integrated toolset platform, referred to as Flexible Artifact Change Traceability Support IDE (FACTS-IDE). Our approach uses an abstract semantic graph (ASG) to extract, model, and represent code change semantics (SETS), create connections between changes and reasons, and enable visualizations for understanding changes. We evaluate our techniques and gather data from interviews, user experiments, and performance measures of the tool. Results indicate that FACTS-IDE aids maintenance developers in capturing traceability links, thus enabling improved understanding of code changes.

Tuesday, May 31st

Anand Joglekar
Chair: Dr. Wooyoung Kim
Candidate: Master of Science in Computer Science & Software Engineering
10:00 AM; DISC 464
Project NeMo
Discovering Motifs in Biological Networks
Using Simple MapReduce

A network motif is a frequent and unique subgraph pattern in a network, which is applied to understand many real-biological problems. However, identification of network motif is computationally intensive, involving three subtasks: enumeration, graph labeling, and statistical uniqueness testing. Therefore, the project in this paper speeds up discovery of network motifs by parallelizing network motif discovery algorithms. Parallelization of network motif discovery was inevitable, since serial discovery is infeasible in large networks, or for large motif sizes. Multiple parallelization solutions have been developed so far, including the most recent methods, such as: iterative MapReduce-based Network Motif detection using Hadoop, and Agent based discovery using MASS library. The work reported on in this paper uses a simple MapReduce architecture based on Spark and Hadoop, running on publicly available Google Compute Cloud. The enumeration subtask is complete, showing near linear performance gain with each additional computing node, while iterative MapReduce in Hadoop method fails to show linear performance gain. We will continue to implement the remaining subtasks, so that the simple MapReduce in Spark network motif discovery program is easily accessible in near future.

Minaashi Kalyanaraman
Chair: Mark Kochanski, M.S.
Candidate: Master of Science in Computer Science & Software Engineering
2:30 PM; DISC 464
Customer Satisfaction in Software Industry

Changing requirements and fast paced product delivery has always been a challenge to measure customer satisfaction in software industry. The strong association of customer satisfaction with quality influences the methods used to measure customer satisfaction. This research project suggests recommendations to improve customer satisfaction in existing software development processes, by analyzing the data collected from 30 participants from 10 software companies. This project also summarizes the different methods used to measure quality and customer satisfaction. Thus by analyzing the impact of customers on software development processes, this project captures the relationship between customers and software industry.

Thursday, June 2nd

Mahmood Khadeer
Chair: Dr. Marc Dupuis
Candidate: Master of Science in Cyber Security Engineering
1:30 PM; DISC 464
A Quantifiable Security Compliance Measurement System for Internet of Things (IoT)

The goal of this paper is to provide a quantitative security compliance  measurement system for consumer IoT devices in a simplistic way. The growing explosion of IoT devices and increasing adoption by consumers makes it more critical that these devices are secure. Specially for consumers there is no simplistic way to find out the security of the IoT devices they are planning to purchase or using. The developed framework uses the security requirements from  Online Trust Alliance (OTA) augmented with critical design and development requirements to assess the complete security status of a consumer IoT device. The framework uses a 1000 point rating system to rate each device on the basis of security validation and verification. The ratings for each requirements are validated by industry experts. A pilot dashboard is developed to show the proof of concept of such an approach for consumers to see the security status of a device. Additional consideration is done to show the business potential of such a service to the growing space of IoT vendors and consumers.

Friday, June 3rd

Dexter Hu
Chair: Dr. Kelvin Sung
Candidate: Master of Science in Computer Science & Software Engineering
10:00 AM; DISC 464
Investigation into GUI Support for a Web-Based Game Engine

A game engine is an Application Programming Interface (API) designed to support game development. Typically, advanced programming skills and knowledge of the back-end API structure are required to build games with these. A well-designed graphical user interface (GUI) to a game engine would remove the need for these requirements and allow designers without in-depth technical background to develop games using the game engine through the GUI. In a web-based environment, such a GUI would allow designers to build and play games from anywhere on the internet. This project, based on the existing web-based GTCS game engine, investigates the technical requirements and factors affecting the usability of the GUI frontend. Our study analyzed GUI modules of existing, popular game engine GUI systems, identified web-based technologies suitable for building the modules, developed prototypes, and integrated the prototypes into a functional web-based GUI frontend for the GTCS game engine. Our results demonstrated feasibility and are the pathway for a user-friendly GUI system of a web-based game engine.

Monday, June 6th

Jebediah Pavleas
Chair: Dr. Kelvin Sung
Candidate: Master of Science in Computer Science & Software Engineering
3:00 PM; DISC 464
Improving Eye Gaze Wheelchair Safety and Usability

Disabled electric wheelchair users with severe physical limitations, such as those with ALS, often require a more elaborate control interface than the traditional joystick. Furthermore, as a disease such as ALS progresses the only voluntary movement a person may control is their eyes. An eye gaze or eye tracking wheelchair affords these users the ability to drive their wheelchair freely by using a sensor to track the user's eye position as they focus on a screen in front of them. With the user's focus directed at the screen rather than the environment around them providing the user with a safe and positive experience while operating system becomes a primary objective. This project explorers the design and implementation of several features created in order to improve the safety and usability of an existing eye gaze wheelchair. These features include a camera system to provide a 360-degree view around the user's wheelchair, user interface optimizations to improve the usability of the system, and a semi-autonomous drive mode to assist the user when operating the chair.

Tuesday, June 7th

Victor De Lima Soares
Chair: Dr. Brent Lagesse
Candidate: Master of Science in Cyber Security Engineering
3:30 PM; DISC 464
TrueNTHConnect: TrueNTH OAuth Role Based Permission System

The objective of this project was the development of an OAuth based authentication, identification and authorization system that was destined to be incorporated into Liferay 6.2 installations, providing portals hosted by the platform with the capability of transparently recognizing accounts managed by a remote central authority. As part of the broader project, denominated TrueNTH, a Movember Foundation initiative, this endeavor sought the development of integrated security frameworks, named TrueNTHConnect, in alignment with the efforts made by the Clinical Informatics Research Group at the University of Washington. Our ultimate goal was to provide a safe environment where information could be shared to improve cancer patients’ life experiences.

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