Matching Items (2)
Filtering by
- All Subjects: Data Mining
- Status: Published
Description
This dissertation addresses the research challenge of developing efficient new methods for discovering useful patterns and knowledge in large volumes of electronically collected spatiotemporal activity data. I propose to analyze three types of such spatiotemporal activity data in a methodological framework that integrates spatial analysis, data mining, machine learning, and geovisualization techniques. Three different types of spatiotemporal activity data were collected through different data collection approaches: (1) crowd sourced geo-tagged digital photos, representing people's travel activity, were retrieved from the website Panoramio.com through information retrieval techniques; (2) the same techniques were used to crawl crowd sourced GPS trajectory data and related metadata of their daily activities from the website OpenStreetMap.org; and finally (3) preschool children's daily activities and interactions tagged with time and geographical location were collected with a novel TabletPC-based behavioral coding system. The proposed methodology is applied to these data to (1) automatically recommend optimal multi-day and multi-stay travel itineraries for travelers based on discovered attractions from geo-tagged photos, (2) automatically detect movement types of unknown moving objects from GPS trajectories, and (3) explore dynamic social and socio-spatial patterns of preschool children's behavior from both geographic and social perspectives.
ContributorsLi, Xun (Author) / Anselin, Luc (Thesis advisor) / Koschinsky, Julia (Committee member) / Maciejewski, Ross (Committee member) / Rey, Sergio (Committee member) / Griffin, William (Committee member) / Arizona State University (Publisher)
Created2012
Spatial-Temporal Analysis of Barrett Freshmen 2007-2012: Source Area Analysis and Poisson Regression
Description
In order to help enhance admissions and recruiting efforts, this longitudinal study analyzed the geographic distribution of matriculated Barrett freshmen from 2007-2012 and sought to explore hot and cold spot locations of Barrett enrollment numbers using geographic information science (GIS) methods. One strategy involved   weighted mean center and standard distance analyses for each year of data for non-resident (out-of-state) freshmen home zip codes. Another strategy, a Poisson regression model, revealed recruitment "hot and cold spots" across the U.S. to project the expected counts of Barrett freshmen by zip code. This projected count served as a comparison for the actual admissions data, where zip codes with over and under predictions represented cold and hot spots, respectively. The mean center analysis revealed a westward shift from 2007 to 2012 with similar distance dispersions. The Poisson model projected zero-student zip codes with 99.2% accuracy and non-zero zip codes with 73.8% accuracy. Norwalk, CA (90650) and New York, NY (10021) represented the top out-of-state cold spot zip codes, while the model indicated that Chandler, AZ (85249) and Queen Creek, AZ (85242) had the most in-state potential for recruitment. The model indicated that more students have come from Albuquerque, NM (87122) and Aurora, CO (80015) than anticipated, while Phoenix, AZ (85048) and Tempe, AZ (85284) represent in-state locations with higher correlations between the variables included, especially regarding distance decay, and the than expected numbers of freshmen. The regression also indicated the existence of strong likelihood of attracting Barrett students.
ContributorsKostanick, Megan Elizabeth (Author) / Rey, Sergio (Thesis director) / Dorn, Ron (Committee member) / Koschinsky, Julia (Committee member) / Barrett, The Honors College (Contributor) / School of Geographical Sciences and Urban Planning (Contributor) / School of Politics and Global Studies (Contributor)
Created2013-05