Files in this item



application/pdfLV-DISSERTATION-2018.pdf (4MB)
(no description provided)PDF


Title:Semantic process analysis, context-aware information retrieval, and sentiment analysis for supporting transportation project environmental review
Author(s):Lv, Xuan
Director of Research:El-Gohary, Nora
Doctoral Committee Chair(s):El-Gohary, Nora
Doctoral Committee Member(s):Liu, Liang; Zhai, ChengXiang; El-Rayes, Khaled; Golparvar-Fard, Mani
Department / Program:Civil & Environmental Eng
Discipline:Civil Engineering
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):semantic analysis, context-aware information retrieval, sentiment analysis, transportation project environmental review
Abstract:According to the National Environmental Policy Act (NEPA), transportation projects are required to go through an environmental review process to evaluate their impact on the environment. However, the Transportation Project Environmental Review process (TPER), has long been “criticized for resulting in frequent delays in the development of important projects designed to improve the safety and operating conditions of a region's transportation system” (FHWA 2013); the time to complete the environmental review process for large-scale transportation projects nearly tripled since the 1970s (Clark and Canter 1997; Barberio et al. 2008a; Venner Consulting et al. 2012). Based on a number of studies (e.g., Mallett and Luther 2011; Cambridge Systematics, Inc. 2011; Keck et al. 2010; FHWA 2016) conducted to identify the constraints for accelerating the TPER process, three primary causes of process inefficiencies were identified: (1) NEPA and transportation project planning processes are not streamlined; (2) transportation practitioners have limited ability to find the right information, at the right time to support mission-critical analyses (Spy Pond Parteners et al. 2009): and (3) there is late identification of stakeholder concerns and support levels. Towards addressing these three problems, this thesis aims to enhance the efficiency of the TPER process through (1) discovering the practices that should be implemented to integrate the NEPA process into the transportation planning process in a manner to ensure both the efficiency of project development and compliance with NEPA; (2) developing context-aware information retrieval methods to support the search and retrieval of relevant textual information in the TPER domain; and (3) developing stakeholder opinion mining methods to identify potential concerns and stakeholder support levels early in the project development process. Accordingly, the thesis includes eight primary research tasks: (1) conducting a comprehensive literature review; (2) analyzing existing processes and identifying successful integration practices for integrating NEPA into transportation planning processes for large-scale highway projects in Illinois; (3) developing a semantic annotation method and algorithm for supporting context-aware information retrieval in the TPER domain; (4) developing a semantic, context-aware information retrieval method and algorithm for retrieving relevant information for supporting the TPER process; (5) developing a stakeholder opinion extraction method and algorithm for automatically extracting subject, concern, and opinion expressions from stakeholder comments on large-scale highway projects to support aspect-level stakeholder opinion mining in the TPER domain; (6) developing a stakeholder opinion classification method and algorithm for classifying the extracted subject, concern, and opinion expressions to support aspect-level opinion mining in the TPER domain; (7) developing a sentence-level opinion mining method and algorithm for classifying comment sentences on large-scale highway projects; and (8) conducting case studies to analyze the differences and similarities among different stakeholder groups in terms of concerns and support levels. All proposed methods and algorithms were tested and evaluated, and the results of these evaluations are presented in the thesis. The thesis also discusses the limitations and recommendations for future research.
Issue Date:2018-04-20
Rights Information:Copyright 2018 Xuan Lv
Date Available in IDEALS:2018-09-04
Date Deposited:2018-05

This item appears in the following Collection(s)

Item Statistics