1,721,037 research outputs found
Applications of balance optimization subset selection
Full text linkBalance Optimization Subset Selection (BOSS) is a framework designed to be used for causal inference on observational data. The theoretical foundation for the BOSS framework has been provided in the literature; this thesis aims to provide some examples of the practical value of BOSS by using it on two problems. The first application is using BOSS to determine a subset of users who would be suitable targets for marketing efforts, and the second application is using BOSS to identify potential first-round upsets in the NCAA basketball tournament. Finally, this thesis delves into another area of college basketball and attempts to model the process of the NCAA tournament selection committee using a decision tree.Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2018-08-01The student, Shouvik Dutta, accepted the attached license on 2016-07-19 at 12:51.The student, Shouvik Dutta, submitted this Thesis for approval on 2016-07-19 at 12:54.This Thesis was approved for publication on 2016-07-21 at 14:08.DSpace SAF Submission Ingestion Package generated from Vireo submission #10013 on 2016-11-10 at 12:21:00Made available in DSpace on 2016-11-10T18:27:07Z (GMT). No. of bitstreams: 2
DUTTA-THESIS-2016.pdf: 645024 bytes, checksum: 2db51a406961fe95b3cc42b8d0e55feb (MD5)
LICENSE.txt: 4210 bytes, checksum: c742995ea6b0e1b88acbf8698e6a703a (MD5)
Previous issue date: 2016-07-21Embargo set by: Seth Robbins for item 95292
Lift date: 2018-11-10T18:28:02Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemLimited Restriction Lifted for Item 95292 on 2018-11-11T10:15:19Z
Five decades of healthcare simulation
No full textIn this paper we have not attempted to produce any kind of systematic review of simulation in healthcare to compete with the dozen (at least) excellent and comprehensive survey papers on this topic that already exist. We begin with a glance back at the early days of Wintersim, but then proceed, in line with the theme of this special track, to reflect on general developments in healthcare simulation over the years from our own personal perspectives. We include some memories and reflections by several pioneers in this area, both academics and healthcare practitioners, on both sides of the Atlantic. We also asked four current simulation modelers, who all specialize in healthcare applications but from very diverse perspectives, to reflect on their experiences. We endeavor to identify some common or recurring themes across the years, and end with a glimpse into the future.</p
Modeling the winning seed distribution of the NCAA basketball tournament
Full text linkThe National Collegiate Athletic Association's (NCAA) men's division I college basketball tournament is an annual competition that draws widespread attention in the United States. Estimating the outcome of each game is a popular activity undertaken by numerous websites, fans, and more recently, academic researchers. There has been a surge of interest in proposing mathematical methods to model the tournament's results and pick the winners of future games. This thesis analyzes the results of the NCAA basketball tournament since 1985 and proposes several models to capture the winning seed distribution in each round.
The Exponential Model estimates the winning probability of each team by modeling the time between a team's successive winnings in a round as an exponential random variable. The Exponential Model estimates a zero probability for events that have not occurred in the training data set. The Markov Model solves this limitation by defining a Markov chain that incorporates each team's winnings in prior rounds to estimate its winning probability. Results of these two models are validated using a chi-squared goodness of fit test.
The Power Model, which is an intelligent tool for generating brackets of winners, quantifies the relative strength of each match-up in a round as a power function of the teams' seed numbers, with the exponent estimated using the historical results. The main problem of the Power Model is the data complications that are generally caused by the small size of the training data set, especially in later rounds. The Position and Upset Models solve this problem by representing the tournament's games as a binary sequence and estimating the outcome of each game based on the teams' performance in the similar game.
While generating a bracket in a forward direction from the first to the last round propagates the incorrect picks through the tournament, correctly picking the winners in later rounds automatically fills the bracket for several games in earlier rounds. This motivates developing bidirectional models that pick the winners based on a combination of models in forward and backward directions. The Power, Position, Upset, and bidirectional models are assessed based on the aggregate performance of millions of brackets for the five most recent tournaments (2012-2016).
The proposed models allow one to estimate the likelihoods of different seed combinations by applying the estimated winning seed distributions, which accurately summarize the seeds' aggregate performance and provide a deeper understanding of the uncertainty in the games' outcomes.Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2017-02-28 without embargo termsThe student, Arash Khatibi, accepted the attached license on 2016-11-22 at 10:42.The student, Arash Khatibi, submitted this Thesis for approval on 2016-11-22 at 10:49.This Thesis was approved for publication on 2016-11-23 at 09:58.DSpace SAF Submission Ingestion Package generated from Vireo submission #10295 on 2017-02-28 at 14:51:46Made available in DSpace on 2017-03-01T15:49:02Z (GMT). No. of bitstreams: 2
KHATIBI-THESIS-2016.pdf: 251783 bytes, checksum: f79936220890914e8bac1f11fc370d01 (MD5)
LICENSE.txt: 4210 bytes, checksum: 0008c829a88649141a17ec03b51b6684 (MD5)
Previous issue date: 2016-11-2
Essays on the relationship between public transit usage and obesity
Full text linkMy dissertation consists of two essays which analyze the impact of public transit usage on obesity.
Chapter 1 introduces the backgrounds of this field and layout the general framework of this thesis work.
Chapter 2 conducts a cross sectional study on the impact of county population level public transit usage on obesity rates. Since the obese population may have different commuting preference in comparison to non-obese population, one can over or under estimate this effect if these preference differences are not properly controlled. This study adopts an instrumental regression approach to implicitly control for the possible selection bias due to different commuting preferences among different populations. The 2009 health data from the Behavioral Risk Factor Surveillance System (BRFSS) and transportation data from the 2009 National Household Travel Survey (NHTS) are aggregated and matched at the county level. Measures of county level public transit accessibility and vehicle ownership rates are chosen as instrumental variables to implicitly control for unobservable commuting preferences. The model suggests that a one percent increase in county population usage of public transit is associated with a 0.287 percent decrease in county population obesity rate at the alpha=0.01 statistical significance level, when commuting preferences, amount of non-travel physical activity, health resource and distribution of income are fixed. This study provides empirical support for the effectiveness of encouraging public transit usage as an intervention strategy for obesity.
Chapter 3 presents a longitudinal study on this topic. Annual health data from the Behavioral Risk Factor Surveillance System (BRFSS) and transportation data from the National Household Travel Survey (NHTS) were aggregated and matched at the county level, to create a panel data set with 229 counties (from 45 states) across two time periods, 2001 and 2009. Possible confounding variables such as amount of leisure time physical activity, health care coverage and distribution of income are explicitly controlled. All time-invariant county level heterogeneities are implicitly controlled using first difference estimators. This study shows that making frequent public transit commuting possible in a county can effectively decrease the county obesity rate. Specifically, a one percent emergence of frequent public transit riders in a county population is estimated to decrease the county population obesity rate by 0.18% or more. This result supports findings in previous research that the extra amount of physical activities involved in public transit usage can have a statistically significant impact on obesity. In addition, this study also provides empirical evidence for the effectiveness of encouraging public transit usage as a public health intervention for obesity.
Chapter 4 concludes this thesis work as well as postulates directions for future study.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2019-08-01The student, Zhaowei She, accepted the attached license on 2017-07-19 at 13:19.The student, Zhaowei She, submitted this Thesis for approval on 2017-07-19 at 13:32.This Thesis was approved for publication on 2017-07-19 at 15:32.DSpace SAF Submission Ingestion Package generated from Vireo submission #11534 on 2018-03-02 at 13:02:43Made available in DSpace on 2018-03-02T19:59:47Z (GMT). No. of bitstreams: 2
SHE-THESIS-2017.pdf: 271866 bytes, checksum: ba50cb7683eb336cb124598b89b2eaa0 (MD5)
LICENSE.txt: 4208 bytes, checksum: c6051d6cb2099edb9e17bafe3edfe842 (MD5)
Previous issue date: 2017-07-19Embargo set by: Seth Robbins for item 105082
Lift date: 2020-03-02T19:59:52Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 105082
Lift date: 2020-03-02T20:02:46Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 105082 on 2020-03-03T10:15:11Z
Variations of online bipartite matching
Full text linkThe Online Bipartite Matching Problem is a well-studied problem in theoretical computer science that models several real-world applications including online investment, kidney transplantation, aviation security passenger screening, and enhanced Ebola entry screening. However, the original version of the problem is too simplistic to cover many real-world applications. Therefore it is common to consider variations of the problem that more closely model the target application.
This thesis considers two variations of the problem motivated by aviation security passenger screening. The first, known as the online total bipartite matching problem, is a variation in which jobs must be assigned to some worker regardless of whether or not it is adjacent to an available worker. Tight upper and lower bounds are given for the general version of this problem, along with 1-competitive algorithm for a special case of the problem.
The second variation begins with the well-known Stochastic Sequential Assignment Problem, which is a variation of the Online Bipartite Matching problem in which edge weights are calculated as the product of a job value and worker value. It extends this to the Reusable Sequential Stochastic Assignment Problem, in which workers can be reused after they finish processing a job. We consider both the stochastic and random arrival model and provide algorithms with constant approximation ratios when job lengths are constant.Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2021-09-16 without embargo termsThe student, Meghan Kelley, accepted the attached license on 2021-04-21 at 16:24.The student, Meghan Kelley, submitted this Thesis for approval on 2021-04-21 at 16:28.This Thesis was approved for publication on 2021-04-23 at 16:31.DSpace SAF Submission Ingestion Package generated from Vireo submission #16479 on 2021-09-16 at 16:46:48Made available in DSpace on 2021-09-17T01:11:11Z (GMT). No. of bitstreams: 2
KELLEY-THESIS-2021.pdf: 343279 bytes, checksum: beb3ebb2740d6278c3b0e0678c52173a (MD5)
LICENSE.txt: 4210 bytes, checksum: 08fc30b90726124ba586ba23539ef41a (MD5)
Previous issue date: 2021-04-2
The Design and Analysis of Pediatric Vaccine Formularies: Theory and Practice
No full textThe first problem models a general childhood immunization schedule to design a vaccine formulary that minimizes the cost of fully immunizing a child. The second problem models a general childhood immunization schedule to design a vaccine formulary that safely immunizes a child against several infectious diseases by restricting or limiting extraimmunization (i.e., extra doses of vaccine). These problems are vitally important since the cost of vaccinating a child contributes to the underimmunization of children, and extraimmunization poses biological risks, amplifies philosophical concerns with vaccination, and creates an unnecessary economic burden on society. These models are rigorously analyzed and several algorithms---both exact and heuristic---are presented. Furthermore, a computational comparison of these algorithms is presented for the 2006 Recommended Childhood Immunization Schedule as well as several randomly generated childhood immunization schedules. The third problem combines the first two problems by modeling a general childhood immunization schedule to design a vaccine formulary that minimizes the cost of fully immunizing a child while restricting or limiting extraimmunization. The results reported here provide both fundamental insights to the operations research community as well as practical value for the public health community.Made available in DSpace on 2015-09-28T15:37:19Z (GMT). No. of bitstreams: 2
license.txt: 4848 bytes, checksum: 96035ab3f5e1c23cc7138a224ce498bd (MD5)
3242861.pdf: 5047706 bytes, checksum: 21c47e7cc167badbfb088641791b34d5 (MD5)
Previous issue date: 2006Embargo set by: Seth Robbins for item 88362
Lift date: Forever
Reason: Restricted to the U of I community idenfinitely during batch ingest of legacy ETDsRestricted to the U of I community idenfinitely during batch ingest of legacy ETDsU of I Only143 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2006
Modeling and analyzing the NCAA Men’s Division I Basketball Tournament
Full text linkThe National Collegiate Athletic Association (NCAA) Men's Division I Basketball Tournament is an annually held basketball tournament between top universities throughout the United States. Along with attracting tens of millions of viewers, the event has become increasingly ingrained in popular culture, with millions attempting to predict the results of the tournament. Naturally, this interest among the general public has sparked similar interest among researchers attempting to statistically model the tournament. This thesis continues these efforts by proposing several methods of estimating the probability distributions of matches. Statistical analysis is conducted to verify these models and various properties of the tournament itself.
There are many challenges to face when developing probabilistic models for this tournament. In particular, the relative scarcity of past data (33 years of past tournaments) combined with the sheer number of possible outcomes (2^63 possible brackets) can make formulating accurate models a daunting task. This thesis proposes the following novel methods of estimating winning probabilities of each match of the tournament. The Position Model estimates winning probability distributions using maximum likelihood estimations based on the position of seeds in the bracket. The Upset Model estimates winning probability distributions using maximum likelihood estimations based on the probability of an upset in any given match. In addition to these two models, this thesis puts forth methods of combining the Position and Upset Model with the Geometric Model proposed by Jacobson et al.
The models proposed in this thesis are verified through the use of various numerical experiments and statistical analysis. In particular, tens of millions of brackets are generated independently at random according to the proposed models. Assessed using a fairly ubiquitous scoring standard, these generated brackets are compared to those submitted by human participants in popular competitions. Further statistical analysis is performed to investigate and support various aspects of these models.Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2018-03-13 without embargo termsThe student, Kevin Li, accepted the attached license on 2017-12-05 at 10:16.The student, Kevin Li, submitted this Thesis for approval on 2017-12-05 at 10:22.This Thesis was approved for publication on 2017-12-06 at 11:19.DSpace SAF Submission Ingestion Package generated from Vireo submission #11842 on 2018-03-13 at 10:10:32Made available in DSpace on 2018-03-13T15:48:54Z (GMT). No. of bitstreams: 2
LI-THESIS-2017.pdf: 311284 bytes, checksum: 48b5aca48dd419defa5c63a954ff0e7c (MD5)
LICENSE.txt: 4205 bytes, checksum: a0195e9608bea9c774ba1113a3c12fa2 (MD5)
Previous issue date: 2017-12-0
The Design and Analysis of Pediatric Vaccine Formularies: Theory and Practice
No full text143 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2006.The first problem models a general childhood immunization schedule to design a vaccine formulary that minimizes the cost of fully immunizing a child. The second problem models a general childhood immunization schedule to design a vaccine formulary that safely immunizes a child against several infectious diseases by restricting or limiting extraimmunization (i.e., extra doses of vaccine). These problems are vitally important since the cost of vaccinating a child contributes to the underimmunization of children, and extraimmunization poses biological risks, amplifies philosophical concerns with vaccination, and creates an unnecessary economic burden on society. These models are rigorously analyzed and several algorithms---both exact and heuristic---are presented. Furthermore, a computational comparison of these algorithms is presented for the 2006 Recommended Childhood Immunization Schedule as well as several randomly generated childhood immunization schedules. The third problem combines the first two problems by modeling a general childhood immunization schedule to design a vaccine formulary that minimizes the cost of fully immunizing a child while restricting or limiting extraimmunization. The results reported here provide both fundamental insights to the operations research community as well as practical value for the public health community.U of I OnlyRestricted to the U of I community idenfinitely during batch ingest of legacy ETD
A Local Search Algorithm Approach to Analyzing the Complexity of Discrete Optimization Problems
No full textThe properties of polynomially computable neighborhood functions are also examined. The global verification of several problems is proven to be NP-complete. These results are extended to show that polynomially computable neighborhood functions have arbitrarily poor local optima.Made available in DSpace on 2015-09-28T15:37:17Z (GMT). No. of bitstreams: 2
license.txt: 4848 bytes, checksum: 96035ab3f5e1c23cc7138a224ce498bd (MD5)
3044043.pdf: 8354499 bytes, checksum: db8df46d5b46177d65aefadc5c48a1e0 (MD5)
Previous issue date: 2002Embargo set by: Seth Robbins for item 88357
Lift date: Forever
Reason: Restricted to the U of I community idenfinitely during batch ingest of legacy ETDsRestricted to the U of I community idenfinitely during batch ingest of legacy ETDsU of I Only178 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2002
Designing Aviation Security Systems: Theory and Practice
No full text122 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2006.Next-generation aviation security systems need not merely be makeshift political solutions for mending complex problems; they can be the result of modeling, analysis, and planning. This dissertation provides a systematic approach for designing and analyzing aviation security systems, which provides insight into the operation of passenger screening systems and guidance for the design of next-generation aviation security systems.U of I OnlyRestricted to the U of I community idenfinitely during batch ingest of legacy ETD
- …
