Matching Items (6)
Filtering by
- All Subjects: deep learning
- All Subjects: spatial statistics
- Genre: Academic theses
- Creators: Li, Wenwen
Description
As urban populations become increasingly dense, massive amounts of new 'big' data that characterize human activity are being made available and may be characterized as having a large volume of observations, being produced in real-time or near real-time, and including a diverse variety of information. In particular, spatial interaction (SI) data - a collection of human interactions across a set of origins and destination locations - present unique challenges for distilling big data into insight. Therefore, this dissertation identifies some of the potential and pitfalls associated with new sources of big SI data. It also evaluates methods for modeling SI to investigate the relationships that drive SI processes in order to focus on human behavior rather than data description.
A critical review of the existing SI modeling paradigms is first presented, which also highlights features of big data that are particular to SI data. Next, a simulation experiment is carried out to evaluate three different statistical modeling frameworks for SI data that are supported by different underlying conceptual frameworks. Then, two approaches are taken to identify the potential and pitfalls associated with two newer sources of data from New York City - bike-share cycling trips and taxi trips. The first approach builds a model of commuting behavior using a traditional census data set and then compares the results for the same model when it is applied to these newer data sources. The second approach examines how the increased temporal resolution of big SI data may be incorporated into SI models.
Several important results are obtained through this research. First, it is demonstrated that different SI models account for different types of spatial effects and that the Competing Destination framework seems to be the most robust for capturing spatial structure effects. Second, newer sources of big SI data are shown to be very useful for complimenting traditional sources of data, though they are not sufficient substitutions. Finally, it is demonstrated that the increased temporal resolution of new data sources may usher in a new era of SI modeling that allows us to better understand the dynamics of human behavior.
A critical review of the existing SI modeling paradigms is first presented, which also highlights features of big data that are particular to SI data. Next, a simulation experiment is carried out to evaluate three different statistical modeling frameworks for SI data that are supported by different underlying conceptual frameworks. Then, two approaches are taken to identify the potential and pitfalls associated with two newer sources of data from New York City - bike-share cycling trips and taxi trips. The first approach builds a model of commuting behavior using a traditional census data set and then compares the results for the same model when it is applied to these newer data sources. The second approach examines how the increased temporal resolution of big SI data may be incorporated into SI models.
Several important results are obtained through this research. First, it is demonstrated that different SI models account for different types of spatial effects and that the Competing Destination framework seems to be the most robust for capturing spatial structure effects. Second, newer sources of big SI data are shown to be very useful for complimenting traditional sources of data, though they are not sufficient substitutions. Finally, it is demonstrated that the increased temporal resolution of new data sources may usher in a new era of SI modeling that allows us to better understand the dynamics of human behavior.
ContributorsOshan, Taylor Matthew (Author) / Fotheringham, A. S. (Thesis advisor) / Farmer, Carson J.Q. (Committee member) / Rey, Sergio S.J. (Committee member) / Nelson, Trisalyn (Committee member) / Arizona State University (Publisher)
Created2017
Description
Big data that contain geo-referenced attributes have significantly reformed the way that I process and analyze geospatial data. Compared with the expected benefits received in the data-rich environment, more data have not always contributed to more accurate analysis. “Big but valueless” has becoming a critical concern to the community of GIScience and data-driven geography. As a highly-utilized function of GeoAI technique, deep learning models designed for processing geospatial data integrate powerful computing hardware and deep neural networks into various dimensions of geography to effectively discover the representation of data. However, limitations of these deep learning models have also been reported when People may have to spend much time on preparing training data for implementing a deep learning model. The objective of this dissertation research is to promote state-of-the-art deep learning models in discovering the representation, value and hidden knowledge of GIS and remote sensing data, through three research approaches. The first methodological framework aims to unify varied shadow into limited number of patterns, with the convolutional neural network (CNNs)-powered shape classification, multifarious shadow shapes with a limited number of representative shadow patterns for efficient shadow-based building height estimation. The second research focus integrates semantic analysis into a framework of various state-of-the-art CNNs to support human-level understanding of map content. The final research approach of this dissertation focuses on normalizing geospatial domain knowledge to promote the transferability of a CNN’s model to land-use/land-cover classification. This research reports a method designed to discover detailed land-use/land-cover types that might be challenging for a state-of-the-art CNN’s model that previously performed well on land-cover classification only.
ContributorsZhou, Xiran (Author) / Li, Wenwen (Thesis advisor) / Myint, Soe Win (Committee member) / Arundel, Samantha Thompson (Committee member) / Arizona State University (Publisher)
Created2019
Description
This thesis encompasses a comprehensive research effort dedicated to overcoming the critical bottlenecks that hinder the current generation of neural networks, thereby significantly advancing their reliability and performance. Deep neural networks, with their millions of parameters, suffer from over-parameterization and lack of constraints, leading to limited generalization capabilities. In other words, the complex architecture and millions of parameters present challenges in finding the right balance between capturing useful patterns and avoiding noise in the data. To address these issues, this thesis explores novel solutions based on knowledge distillation, enabling the learning of robust representations. Leveraging the capabilities of large-scale networks, effective learning strategies are developed. Moreover, the limitations of dependency on external networks in the distillation process, which often require large-scale models, are effectively overcome by proposing a self-distillation strategy. The proposed approach empowers the model to generate high-level knowledge within a single network, pushing the boundaries of knowledge distillation. The effectiveness of the proposed method is not only demonstrated across diverse applications, including image classification, object detection, and semantic segmentation but also explored in practical considerations such as handling data scarcity and assessing the transferability of the model to other learning tasks. Another major obstacle hindering the development of reliable and robust models lies in their black-box nature, impeding clear insights into the contributions toward the final predictions and yielding uninterpretable feature representations. To address this challenge, this thesis introduces techniques that incorporate simple yet powerful deep constraints rooted in Riemannian geometry. These constraints confer geometric qualities upon the latent representation, thereby fostering a more interpretable and insightful representation. In addition to its primary focus on general tasks like image classification and activity recognition, this strategy offers significant benefits in real-world applications where data scarcity is prevalent. Moreover, its robustness in feature removal showcases its potential for edge applications. By successfully tackling these challenges, this research contributes to advancing the field of machine learning and provides a foundation for building more reliable and robust systems across various application domains.
ContributorsChoi, Hongjun (Author) / Turaga, Pavan (Thesis advisor) / Jayasuriya, Suren (Committee member) / Li, Wenwen (Committee member) / Fazli, Pooyan (Committee member) / Arizona State University (Publisher)
Created2023
Description
Integrated water resources management for flood control, water distribution, conservation, and food security require understanding hydrological spatial and temporal trends. Proliferation of monitoring and sensor data has boosted data-driven simulation and evaluation. Developing data-driven models for such physical process-related phenomena, and meaningful interpretability therein, necessitates an inventive methodology. In this dissertation, I developed time series and deep learning model that connected rainfall, runoff, and fish species abundances. I also investigated the underlying explainabilty for hydrological processes and impacts on fish species. First, I created a streamflow simulation model using computer vision and natural language processing as an alternative to physical-based routing. I tested it on seven US river network sections and showed it outperformed time series models, deep learning baselines, and novel variants. In addition, my model explained flow routing without physical parameter input or time-consuming calibration. On the basis of this model, I expanded it from accepting dispersed spatial inputs to adopting comprehensive 2D grid data. I constructed a spatial-temporal deep leaning model for rainfall-runoff simulation. I tested it against a semi-distributed hydrological model and found superior results. Furthermore, I investigated the potential interpretability for rainfall-runoff process in both space and time. To understand impacts of flow variation on fish species, I applied a frequency based model framework for long term time series data simulation. First, I discovered that timing of hydrological anomalies was as crucial as their size. Flooding and drought, when properly timed, were both linked with excellent fishing productivity. To identify responses of various fish trait groups, I used this model to assess mitigated hydrological variation by fish attributes. Longitudinal migratory fish species were more impacted by flow variance, whereas migratory strategy species reacted in the same direction but to various degrees. Finally, I investigated future fish population changes under alternative design flow scenarios and showed that a protracted low flow with a powerful, on-time flood pulse would benefit fish. In my dissertation, I constructed three data-driven models that link the hydrological cycle to the stream environment and give insight into the underlying physical process, which is vital for quantitative, efficient, and integrated water resource management.
ContributorsDeng, Qi (Author) / Sabo, John (Thesis advisor) / Grimm, Nancy (Thesis advisor) / Ganguly, Auroop (Committee member) / Li, Wenwen (Committee member) / Mascaro, Giuseppe (Committee member) / Arizona State University (Publisher)
Created2022
Description
Spatial regression is one of the central topics in spatial statistics. Based on the goals, interpretation or prediction, spatial regression models can be classified into two categories, linear mixed regression models and nonlinear regression models. This dissertation explored these models and their real world applications. New methods and models were proposed to overcome the challenges in practice. There are three major parts in the dissertation.
In the first part, nonlinear regression models were embedded into a multistage workflow to predict the spatial abundance of reef fish species in the Gulf of Mexico. There were two challenges, zero-inflated data and out of sample prediction. The methods and models in the workflow could effectively handle the zero-inflated sampling data without strong assumptions. Three strategies were proposed to solve the out of sample prediction problem. The results and discussions showed that the nonlinear prediction had the advantages of high accuracy, low bias and well-performed in multi-resolution.
In the second part, a two-stage spatial regression model was proposed for analyzing soil carbon stock (SOC) data. In the first stage, there was a spatial linear mixed model that captured the linear and stationary effects. In the second stage, a generalized additive model was used to explain the nonlinear and nonstationary effects. The results illustrated that the two-stage model had good interpretability in understanding the effect of covariates, meanwhile, it kept high prediction accuracy which is competitive to the popular machine learning models, like, random forest, xgboost and support vector machine.
A new nonlinear regression model, Gaussian process BART (Bayesian additive regression tree), was proposed in the third part. Combining advantages in both BART and Gaussian process, the model could capture the nonlinear effects of both observed and latent covariates. To develop the model, first, the traditional BART was generalized to accommodate correlated errors. Then, the failure of likelihood based Markov chain Monte Carlo (MCMC) in parameter estimating was discussed. Based on the idea of analysis of variation, back comparing and tuning range, were proposed to tackle this failure. Finally, effectiveness of the new model was examined by experiments on both simulation and real data.
In the first part, nonlinear regression models were embedded into a multistage workflow to predict the spatial abundance of reef fish species in the Gulf of Mexico. There were two challenges, zero-inflated data and out of sample prediction. The methods and models in the workflow could effectively handle the zero-inflated sampling data without strong assumptions. Three strategies were proposed to solve the out of sample prediction problem. The results and discussions showed that the nonlinear prediction had the advantages of high accuracy, low bias and well-performed in multi-resolution.
In the second part, a two-stage spatial regression model was proposed for analyzing soil carbon stock (SOC) data. In the first stage, there was a spatial linear mixed model that captured the linear and stationary effects. In the second stage, a generalized additive model was used to explain the nonlinear and nonstationary effects. The results illustrated that the two-stage model had good interpretability in understanding the effect of covariates, meanwhile, it kept high prediction accuracy which is competitive to the popular machine learning models, like, random forest, xgboost and support vector machine.
A new nonlinear regression model, Gaussian process BART (Bayesian additive regression tree), was proposed in the third part. Combining advantages in both BART and Gaussian process, the model could capture the nonlinear effects of both observed and latent covariates. To develop the model, first, the traditional BART was generalized to accommodate correlated errors. Then, the failure of likelihood based Markov chain Monte Carlo (MCMC) in parameter estimating was discussed. Based on the idea of analysis of variation, back comparing and tuning range, were proposed to tackle this failure. Finally, effectiveness of the new model was examined by experiments on both simulation and real data.
ContributorsLu, Xuetao (Author) / McCulloch, Robert (Thesis advisor) / Hahn, Paul (Committee member) / Lan, Shiwei (Committee member) / Zhou, Shuang (Committee member) / Saul, Steven (Committee member) / Arizona State University (Publisher)
Created2020
Description
Geographically Weighted Regression (GWR) has been broadly used in various fields to
model spatially non-stationary relationships. Classic GWR is considered as a single-scale model that is based on one bandwidth parameter which controls the amount of distance-decay in weighting neighboring data around each location. The single bandwidth in GWR assumes that processes (relationships between the response variable and the predictor variables) all operate at the same scale. However, this posits a limitation in modeling potentially multi-scale processes which are more often seen in the real world. For example, the measured ambient temperature of a location is affected by the built environment, regional weather and global warming, all of which operate at different scales. A recent advancement to GWR termed Multiscale GWR (MGWR) removes the single bandwidth assumption and allows the bandwidths for each covariate to vary. This results in each parameter surface being allowed to have a different degree of spatial variation, reflecting variation across covariate-specific processes. In this way, MGWR has the capability to differentiate local, regional and global processes by using varying bandwidths for covariates. Additionally, bandwidths in MGWR become explicit indicators of the scale at various processes operate. The proposed dissertation covers three perspectives centering on MGWR: Computation; Inference; and Application. The first component focuses on addressing computational issues in MGWR to allow MGWR models to be calibrated more efficiently and to be applied on large datasets. The second component aims to statistically differentiate the spatial scales at which different processes operate by quantifying the uncertainty associated with each bandwidth obtained from MGWR. In the third component, an empirical study will be conducted to model the changing relationships between county-level socio-economic factors and voter preferences in the 2008-2016 United States presidential elections using MGWR.
model spatially non-stationary relationships. Classic GWR is considered as a single-scale model that is based on one bandwidth parameter which controls the amount of distance-decay in weighting neighboring data around each location. The single bandwidth in GWR assumes that processes (relationships between the response variable and the predictor variables) all operate at the same scale. However, this posits a limitation in modeling potentially multi-scale processes which are more often seen in the real world. For example, the measured ambient temperature of a location is affected by the built environment, regional weather and global warming, all of which operate at different scales. A recent advancement to GWR termed Multiscale GWR (MGWR) removes the single bandwidth assumption and allows the bandwidths for each covariate to vary. This results in each parameter surface being allowed to have a different degree of spatial variation, reflecting variation across covariate-specific processes. In this way, MGWR has the capability to differentiate local, regional and global processes by using varying bandwidths for covariates. Additionally, bandwidths in MGWR become explicit indicators of the scale at various processes operate. The proposed dissertation covers three perspectives centering on MGWR: Computation; Inference; and Application. The first component focuses on addressing computational issues in MGWR to allow MGWR models to be calibrated more efficiently and to be applied on large datasets. The second component aims to statistically differentiate the spatial scales at which different processes operate by quantifying the uncertainty associated with each bandwidth obtained from MGWR. In the third component, an empirical study will be conducted to model the changing relationships between county-level socio-economic factors and voter preferences in the 2008-2016 United States presidential elections using MGWR.
ContributorsLi, Ziqi (Author) / Fotheringham, A. Stewart (Thesis advisor) / Goodchild, Michael F. (Committee member) / Li, Wenwen (Committee member) / Arizona State University (Publisher)
Created2020