東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back to Search results for [ null ]

Switch To: Labeled | MARC Mode | ISBD

Detecting Outliers for Improving the...

Lu, Yi-Chi.

Linked to FindBook

Google Book

Amazon

博客來

Detecting Outliers for Improving the Quality of Incident Duration Prediction.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Detecting Outliers for Improving the Quality of Incident Duration Prediction./
Author:	Lu, Yi-Chi.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2021,
Description:	95 p.
Notes:	Source: Masters Abstracts International, Volume: 83-02.
Contained By:	Masters Abstracts International83-02.
Subject:	Transportation. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28315647
ISBN:	9798534653335

Detecting Outliers for Improving the Quality of Incident Duration Prediction.
Lu, Yi-Chi.

Detecting Outliers for Improving the Quality of Incident Duration Prediction. - Ann Arbor : ProQuest Dissertations & Theses, 2021 - 95 p.

Source: Masters Abstracts International, Volume: 83-02.

Thesis (M.S.)--University of Maryland, College Park, 2021.

This item must not be sold to any third party vendors.

To circumvent the needs of domain expertise and the excessive data for developing a knowledge-based prediction system such as the I-95 incident duration estimation model, this study has developed an efficient transferability analysis method to assess the applicability of adopting the prediction rules from an existing well-developed model to a different highway. The proposed analysis method has considered the common nature of incident response operations and local-specific incident characteristics in assessing the transferability of available knowledge-based rules for estimating the required clearance duration of different types of incidents. Evaluation of the proposed method with the I-695 incident records clearly shows that the prediction model developed with such an effective transferring method can achieve the same level of performance as with the original rule-searching and refinement method.Since most incident records for model development are collected on-line during the emergency incident response process, some of the key data are likely to be misrecorded which inevitably causes many existing models to yield undesirable performance, especially with respect to those incidents with insufficient records or excessive long duration. As such, this study has also developed a two-phase outlier detection process for identifying outliers and removing those viewed as faulty records from the dataset for model calibration and model evaluation. Using the I-695 incident records for a case study, the resulting performance of the proposed two-phase outlier detection process has proved its promising property for filtering faculty data from the incident records prior to the use for model development.

ISBN: 9798534653335Subjects--Topical Terms:

555912
Transportation.
Subjects--Index Terms:

Incident response operations

Detecting Outliers for Improving the Quality of Incident Duration Prediction.
LDR:02894nmm a2200373 4500 001 2283275
005 20211029084536.5
008 220723s2021 ||||||||||||||||| ||eng d
020 $a 9798534653335
035 $a (MiAaPQ)AAI28315647
035 $a AAI28315647
040 $a MiAaPQ $c MiAaPQ
100 1 $a Lu, Yi-Chi. $3 3562211
245 1 0 $a Detecting Outliers for Improving the Quality of Incident Duration Prediction.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2021
300 $a 95 p.
500 $a Source: Masters Abstracts International, Volume: 83-02.
500 $a Advisor: Chang, Gang-Len.
502 $a Thesis (M.S.)--University of Maryland, College Park, 2021.
506 $a This item must not be sold to any third party vendors.
520 $a To circumvent the needs of domain expertise and the excessive data for developing a knowledge-based prediction system such as the I-95 incident duration estimation model, this study has developed an efficient transferability analysis method to assess the applicability of adopting the prediction rules from an existing well-developed model to a different highway. The proposed analysis method has considered the common nature of incident response operations and local-specific incident characteristics in assessing the transferability of available knowledge-based rules for estimating the required clearance duration of different types of incidents. Evaluation of the proposed method with the I-695 incident records clearly shows that the prediction model developed with such an effective transferring method can achieve the same level of performance as with the original rule-searching and refinement method.Since most incident records for model development are collected on-line during the emergency incident response process, some of the key data are likely to be misrecorded which inevitably causes many existing models to yield undesirable performance, especially with respect to those incidents with insufficient records or excessive long duration. As such, this study has also developed a two-phase outlier detection process for identifying outliers and removing those viewed as faulty records from the dataset for model calibration and model evaluation. Using the I-695 incident records for a case study, the resulting performance of the proposed two-phase outlier detection process has proved its promising property for filtering faculty data from the incident records prior to the use for model development.
590 $a School code: 0117.
650 4 $a Transportation. $3 555912
650 4 $a Urban planning. $3 2122922
650 4 $a Information science. $3 554358
650 4 $a Research. $3 531893
650 4 $a Accuracy. $3 3559958
650 4 $a Datasets. $3 3541416
650 4 $a Response time. $3 3562212
650 4 $a Regression analysis. $3 529831
650 4 $a Estimates. $3 3561047
650 4 $a Artificial intelligence. $3 516317
650 4 $a Roads & highways. $3 3555163
650 4 $a Researchers. $3 3560729
650 4 $a Traffic congestion. $3 706812
650 4 $a Statistical analysis. $3 3543751
650 4 $a Principal components analysis. $3 565921
650 4 $a Knowledge. $3 872758
650 4 $a Neural networks. $3 677449
650 4 $a Classification. $3 595585
650 4 $a Probability. $3 518898
650 4 $a Methods. $3 3560391
650 4 $a Algorithms. $3 536374
650 4 $a Decision trees. $3 827433
650 4 $a Neighborhoods. $3 993475
653 $a Incident response operations
653 $a Outlier Detection
653 $a Local-specific incidentcharacteristics
653 $a Clearance duration
690 $a 0709
690 $a 0723
690 $a 0999
690 $a 0800
710 2 $a University of Maryland, College Park. $b Civil Engineering. $3 1025664
773 0 $t Masters Abstracts International $g 83-02.
790 $a 0117
791 $a M.S.
792 $a 2021
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28315647