Dissertation/Thesis Abstract

The author has requested that access to this graduate work be delayed until 2020-04-16. After this date, this graduate work will be available on an open access basis.
Object Ranking Algorithms for Improved Decision Making
by Duggimpudi, Maria Bala, Ph.D., University of Louisiana at Lafayette, 2017, 135; 10684385
Abstract (Summary)

In this dissertation, by using quantification or ranking of an object/individual, we addressed two problems. Firstly, in the past in spatio-temporal outlier detection, there were no algorithms or methods that quantify spatio-temporal outliers. Here, we propose two algorithms that quantify spatio-temporal outliers, namely Spatio-Temporal Behavioral Density-based Clustering of Applications with Noise (ST-BDBCAN) and Approx-ST-BDBCAN. ST-BDBCAN algorithm adopts the proposed, new concept, called Spatio-Temporal Behavioral Outlier Factor (ST-BOF), which is a spatio-temporal extension to LOF. It also uses both spatial and temporal attributes simultaneously to define the context. The Approx-ST-BDBCAN algorithm achieves improved scalability with minimal loss of detection accuracy by partitioning data points for parallel processing. Experimental results on synthetic and Buoy datasets suggest that our proposed algorithms are accurate and computationally efficient. Additionally, a new Outlier Association with Hurricane Intensity Index (OAHII) measure is introduced for quantitative evaluation of the results from buoy dataset. Secondly, we propose an ontology-based architecture for performing semantic data mining for insights. Later, we illustrate the major components of the architecture using Wireless Networks and University Ranking datasets. In addition, algorithms are presented for summarizing performance profiles in the form of rank tables and for extracting insight rules (concrete action plan) from the rank tables. By using this approach, an actionable plan for assisting decision makers can be obtained, as domain knowledge is incorporated in the system. Experimental results on wireless networks and university ranking datasets show that our model provides an optimal action plan.

Indexing (document details)
Advisor: Raghavan, Vijay V.
Commitee: Amini, Mohsen, Chu, Chee-Hung Henry, Gottumukkala, Raju, Maida, Anthony
School: University of Louisiana at Lafayette
Department: Computer Science
School Location: United States -- Louisiana
Source: DAI-B 79/09(E), Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Computer science
Keywords: Algorithms, Decision making, Insight, Ontology-based, Rank, Spatio-temporal outliers
Publication Number: 10684385
ISBN: 9780355854466
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest