Research Talk Schedule

Primary tabs

The Research Talk format will be a Brown Bag Lunch.
Snacks and Beverages provided.
Begin gathering at noon, talk commences at 12:30 in CP 206M

Date of Presentation Presenter(s) Research Focus
07/03/2013

Topic: Measuring Impact of Professional Development Training in Special Needs Instruction using Data Science

Abstract: This project aims to address the problem of measuring Quality of Professional Development (QPD) of teachers working with special needs students in the State of Washington. We will explore QPD survey data, State Funding Allocation data and Student Performance data in this work. Our plan is to use expressive, dynamically generated visualizations to help discover relationships between key attributes within a dataset and among distinct datasets. With insights gained through visualization we use data analysis to confirm our intuitions about correlations in the data and generate data mining models to make predictions about the likelihood of specific educational outcomes. For special needs education the objective of developing such predictive models is to ensure that special needs students meet proficiency targets, thereby continuously increasing the quality of education made available to these students.

07/10/2013

Presenter: Sofie De Clercq (Ghent University)
Topic: Modeling Stable Matching Problems with Answer Set Programming
Abstract. The Stable Marriage Problem (SMP) is a well-known matching problem first introduced and solved by Gale and Shapley (1962). Several variants and extensions to this problem have since been investigated to cover a wider set of applications. Each time a new variant is considered, however, a new algorithm needs to be developed and implemented. As an alternative, in this paper we propose an encoding of the SMP using Answer Set Programming (ASP). Our encoding can easily be extended and adapted to the needs of specific applications. As an illustration we show how stable matchings can be found when individuals may designate unacceptable partners and ties between preferences are allowed. Subsequently, we show how our ASP based encoding naturally allows us to select specific stable matchings which are optimal according to a given criterion. Each time, we can rely on generic and efficient off-the-shelf answer set solvers to find (optimal) stable matchings.

07/17/2013

Presenter: Muaz Mian

Title: "Query Builder: Exploring Big Data in Real-Time"

Abstract

The amount of data is growing and enterprises want to realize the hidden value behind big data. However,
within enterprises the business users of data are often not database engineers. Hence, analyzing and
exploring big data still requires custom tools. Even if a visual custom tool is available such tools

often are not real-time. Users often wait for hours for a query to execute to extract the relevant data
before they can explore the data, but in last decade there has been some groundbreaking developments in
processing and analyzing data. New technologies and database systems have been invented which are faster

than traditional DBMS. The main goal of this project is to design a browser based tool that can act as an
extremely responsive real-time visual query interface. The tool will be able to plug into any backend big
data database technology including traditional databases, NoSQL systems, and columnar databases. The

efficiency will stem from our proposal to perform automated schema inference, and use main-memory
databases. Time permitting we will design a prototypical main-memory data structure for main-memory
database that will address at least a few if not all the challenges we encounter while using existing

main-memory databases in our tests on healthcare big data. This project will be supported in part by the
UWT-Multicare Health System collaboration initiative.

07/24/2013

Presenters: Kiyana Zolfaghar, Deepthi Sistla, Naren Meadem

Title: "Risk-O-Meter: An Intelligent Clinical Risk Calculator for Readmission in Congestive Heart Failure Patients"

Abstract:

Congestive Heart Failure (CHF) is one of the leading causes of hospitalization, and studies show that many of these admissions are readmissions within a short window of time. Identifying CHF patients who are at a greater risk of hospitalization can guide implementation of appropriate plans to prevent these readmissions.  Developing predictive modeling solutions for such disease related risk of readmissions is extremely challenging in healthcare informatics. It involves integration of socio-demographic factors, health conditions, disease parameters, hospital care quality parameters, and a variety of variables specific to health care providers making the task immensely complex. Data extraction and data preprocessing such as feature selection, missing value imputation and data balancing are some of the prominent steps that we focus on to improve prediction outcomes.

In this project, in collaboration with experts from Multicare Health Systems (MHS), we present a system called Risk-O-Meter to predict and analyze readmission risk via data imputation, visualization, predictive modeling, and association rule exploration.  Risk-O-Meter is designed in a way such that it is flexible enough to accept limited or incomplete data inputs, and still manages to predict the risk efficiently. Moreover, along with the risk of readmission calculation, Risk-O-Meter will also suggest a meaningful explanation of data behind the risk of readmission prediction and offers intelligent visualization of pertinent characteristics of the data, based on input value.

07/31/2013

Presenter: Yitao Li

Topic: Challenges and solutions for extracting meta-data such as title, author, reference information, and reference context from academic publications.

This talk is based on Yitao's experience in the first half of the summer internship at Microsoft.

08/07/2013

Presenters: Graduate Students

Topic: Monthly updates on Student Projects.

08/14/2013

Presenter: Dr. Jie Sheng (University of Washington, Tacoma - Institute of Technology)

08/21/2013

Presenter: Siddharth Bhave (University of Washington, Tacoma - Institute of Technology)

Litesprite is building a mobile game platform designed to assist players who suffer from chronic conditions such as diabetes, heart disease, or general anxiety/depression. For the Robert Wood Johnson Foundation: Games to Generate Data Challenge, Litesprite built a casual game to help people, especially adult women, with anxiety/depression. Through the gameplay, a first-of-kind holistic activity log, or patient registry, is created with other important contextual metrics (i.e. time of day, location) that can help assess the underlying issues related to the player’s concerns. This information can be shared with clinicians and caregivers to help understand the player’s underlying challenges and triggers associated with managing anxiety.
Siddharth's task on this project was to make sure that the gaming app talks to a database and writes data logs as the users play the game. His talk will focus on this data integration problem.

08/28/2013

Presenter: Dr Ankur Teredesai (University of Washington, Tacoma - Institute of Technology)

09/04/2013

NO SEMINAR
CWDS will be planning for new school year.

09/11/2013

Presenters: Student Project Updates

09/18/2013

Presenter: Patrick Pow (University of Washington, Tacoma - Institute of Technology)

09/25/2013

Presenter: Dr Senjuti Basu Roy (University of Washington, Tacoma - Institute of Technology)