ASU Electronic Theses and Dissertations
This collection includes most of the ASU Theses and Dissertations from 2011 to present. ASU Theses and Dissertations are available in downloadable PDF format; however, a small percentage of items are under embargo. Information about the dissertations/theses includes degree information, committee members, an abstract, supporting data or media.
In addition to the electronic theses found in the ASU Digital Repository, ASU Theses and Dissertations can be found in the ASU Library Catalog.
Dissertations and Theses granted by Arizona State University are archived and made available through a joint effort of the ASU Graduate College and the ASU Libraries. For more information or questions about this collection contact or visit the Digital Repository ETD Library Guide or contact the ASU Graduate College at gradformat@asu.edu.
Despite years of research, there are still some unsolved problems on semantic attribute learning. First, real-world applications usually involve hundreds of attributes which requires great effort to acquire sufficient amount of labeled data for model learning. Second, existing attribute learning work for visual objects focuses primarily on images, with semantic analysis on videos left largely unexplored.
In this dissertation I conduct innovative research and propose novel approaches to tackling the aforementioned problems. In particular, I propose robust and accurate learning frameworks on both attribute ranking and prediction by exploring the correlation among multiple attributes and utilizing various types of label information. Furthermore, I propose a video-based skill coaching framework by extending attribute learning to the video domain for robust motion skill analysis. Experiments on various types of applications and datasets and comparisons with multiple state-of-the-art baseline approaches confirm that my proposed approaches can achieve significant performance improvements for the general attribute learning problem.
First, while an increasing amount of investigation has been done in this important area, most existing work concentrates on efficiency instead of search quality and may fail to deliver high quality results from semantic perspectives. Majority of the existing work generates minimal sub-graph results that are oblivious to the entity and relationship semantics embedded in the data and in the user query. There are also studies that define results to be subtrees or subgraphs that contain all query keywords but are not necessarily ``minimal''. However, such result construction method suffers from the same problem of semantic mis-alignment between data and user query. In this work the semantics of how to {\em define} results that can capture users' search intention and then the generation of search intention aware results is studied.
Second, most existing research is incapable of handling large-scale structured data. However, as data volume has seen rapid growth in recent years, the problem of how to efficiently process keyword queries on large-scale structured data becomes important. MapReduce is widely acknowledged as an effective programming model to process big data. For keyword query processing on data graph, first graph algorithms which can efficiently return query results that are consistent with users' search intention are proposed. Then these algorithms are migrated to MapReduce to support big data. For keyword query processing on schema graph, it first transforms a keyword query into multiple SQL queries, then all generated SQL queries are run on the structured data. Therefore it is crucial to find the optimal way to execute a SQL query using MapReduce, which can minimize the processing time. In this work, a system called SOSQL is developed which generates the optimal query execution plan using MapReduce for a SQL query $Q$ with time complexity $O(n^2)$, where $n$ is the number of input tables of $Q$.
This thesis studies the current methodologies in predictive visual analytics. It first defines the scope of predictive analytics and presents a predictive visual analytics (PVA) pipeline. Following the proposed pipeline, a predictive visual analytics framework is developed to be used to explore under what circumstances a human-in-the-loop prediction process is most effective. This framework combines sentiment analysis, feature selection mechanisms, similarity comparisons and model cross-validation through a variety of interactive visualizations to support analysts in model building and prediction. To test the proposed framework, an instantiation for movie box-office prediction is developed and evaluated. Results from small-scale user studies are presented and discussed, and a generalized user study is carried out to assess the role of predictive visual analytics under a movie box-office prediction scenario.
Data integration involves the reconciliation of data from diverse data sources in order to obtain a unified data repository, upon which an end user such as a data analyst can run analytics sessions to explore the data and obtain useful insights. Supervised Machine Learning (ML) for data integration tasks such as ontology (schema) or entity (instance) matching requires several training examples in terms of manually curated, pre-labeled matching and non-matching schema concept or entity pairs which are hard to obtain. On similar lines, an analytics system without predictive capabilities about the impending workload can incur huge querying latencies, while leaving the onus of understanding the underlying database schema and writing a meaningful query at every step during a data exploration session on the user. In this dissertation, I will describe the human-in-the-loop Machine Learning (ML) systems that I have built towards data integration and predictive analytics. I alleviate the need for extensive prior labeling by utilizing active learning (AL) for dataintegration. In each AL iteration, I detect the unlabeled entity or schema concept pairs that would strengthen the ML classifier and selectively query the human oracle for such labels in a budgeted fashion. Thus, I make use of human assistance for ML-based data integration. On the other hand, when the human is an end user exploring data through Online Analytical Processing (OLAP) queries, my goal is to pro-actively assist the human by predicting the top-K next queries that s/he is likely to be interested in. I will describe my proposed SQL-predictor, a Business Intelligence (BI) query predictor and a geospatial query cardinality estimator with an emphasis on schema abstraction, query representation and how I adapt the ML models for these tasks. For each system, I will discuss the evaluation metrics and how the proposed systems compare to the state-of-the-art baselines on multiple datasets and query workloads.
In this study, I aim to achieve multimodal brain image fusion by referring to some intrinsic properties of data, e.g. geometry of embedding structures where the commonly used image features reside. Since the image features investigated in this study share an identical embedding space, i.e. either defined on a brain surface or brain atlas, where a graph structure is easy to define, it is straightforward to consider the mathematically meaningful properties of the shared structures from the geometry perspective.
I first introduce the background of multimodal fusion of brain image data and insights of geometric properties playing a potential role to link different modalities. Then, several proposed computational frameworks either using the solid and efficient geometric algorithms or current geometric deep learning models are be fully discussed. I show how these designed frameworks deal with distinct geometric properties respectively, and their applications in the real healthcare scenarios, e.g. to enhanced detections of fetal brain diseases or abnormal brain development.
In addition, to evaluate the prediction power of this approach, another important aspect of personalized medicine was addressed: the prediction of medication usage through the identification of risk groups. During the prediction process, the health information from Twitter timeline, such as diseases, symptoms, treatments, effects, and etc., is summarized by the topic modelling processes and the summarization results is used for prediction. Dimension reduction and topic similarity measurement are integrated into this framework for timeline classification and prediction. This work could be applied to provide guidelines for FDA drug risk categories. Currently, this process is done based on laboratory results and reported cases.
Finally, a multi-dimensional text data warehouse (MTD) to manage the output from the topic modelling is proposed. Some attempts have been also made to incorporate topic structure (ontology) and the MTD hierarchy. Results demonstrate that proposed methods show promise and this system represents a low-cost approach for drug safety early warning.