Date of Publication

12-14-2019

Document Type

Master's Thesis

Degree Name

Master of Science in Computer Science

Subject Categories

Computer Sciences

College

College of Computer Studies

Department/Unit

Software Technology

Thesis Adviser

Joel Ilao

Defense Panel Chair

Neil Patrick Del Gallego

Defense Panel Member

Edgar Vallar
Joel Ilao

Abstract/Summary

Abstract In the National Capital Region (NCR) of the Philippines, PM2.5 levels are 70%higher than what is considered safe by the World Health Organization’s (WHO) air quality guidelines. One main cause of this is the pervasive use and the increasing number of vehicles in the region. In other countries, monitoring stations and emissions models have aided air quality planning and policy-making. However, acquiring and maintaining air quality monitoring stations are costly. On the other hand, the existing vehicle emissions model for estimating emission quantities requires in-depth adjustments and calibration of the model. As such, this study explored and developed a low-cost alternative system to estimate ambient air quality in line with the Philippine land transport system conditions. The system accepts vehicle counts and meteorological information as input. It can predict PM2.5 levels by applying statistical and machine learning techniques. Among the six models created in this study, LSTM and SVR with time lags produced lower errors and higher correlation with the actual PM2.5 levels. Additionally, the system also generates a dynamic schematic visualization of the inferred PM2.5 levels and the model’s performance.

Abstract Format

html

Language

English

Format

Electronic

Accession Number

CDTG008019

Keywords

Information storage and retrieval systems—Air quality; Air quality—Philippines—Metro Manila; Air—Pollution—Philippines—Metro Manila—Measurement

Upload Full Text

wf_yes

Embargo Period

2-8-2023

Download

Share

COinS