(For USM Staff/Student Only)

EngLib USM > Ω School of Chemical Engineering >

Feature selection and model prediction of air quality using pm2.5

Feature selection and model prediction of air quality using pm2.5 / Sharon Ding Tiew Kui
Kajian ini adalah untuk menjana model peramalan rangkaian neural suap depan (FANN) yang sesuai untuk meramalkan kualiti udara dengan menggunakan PM2.5. Kini, Malaysia masih belum mempunyai model peramalan untuk kepekatan PM2.5. Jadi, dengan model peramalan yang dijana, kepekatan PM2.5 dalam udara dapat diramalkan dengan menggunakan parameter meteologi. Kaedah utama yang diselidik dalam kajian ini ialah bilangan neuron lapisan tersembunyi. Prestasi model peramalan telah dianalisa dan dinilai dengan menggunakan nilai min ralat kuasa dua (MSE) dan pekali penentuan (R2). Dengan menambahkan bilangan neuron dalam lapisan tersembunyi, nilai MSE dapat dikurangkan manakala nilai R2 ditambahkan. 10 neuron lapisan tersembunyi memberikan prestasi yang terbaik antara bilangan neuron yang diselidik. Oleh sebab prestasi model peramalan yang rendah, pemilihan ciri telah diperkenalkan untuk menyingkirkan parameter yang tidak berkaitan dalam set data. Hutan rawak (RF) telah ditumbuhkan dengam 200 pokok regresi untuk menentukan peramal yang paling penting. Peramal yang kurang penting telah disingkirkan daripada peramal yang lain. Dengan penyingkiran parameter yang tidak berkaitan, kejituan model peramalan telah dipertingkatkan dengan peningkatan prestasi model. Selain itu, keserasian model peramalan juga dapat dikurangkan dengan mengurangkan latihan masa model peramalan. Peramal yang disingkirkan oleh pemilihan ciri dalam kajian ini ialah tekanan, titik embun, curah hujan setiap jam dan curah hujan kumulatif. Maka, jelasnya bahawa prestasi model peramalan dengan pemilihan ciri adalah lebih baik daripada prestasi model peramalan tanpa pemilihan ciri. _______________________________________________________________________________________________________ This study was to develop a feed-forward artificial neural network (FANN) prediction model to predict the air quality using PM2.5. Currently, Malaysia does not have any prediction model for concentration of PM2.5. Thus, with the prediction model developed, the concentration of PM2.5 in air can be predicted by using meteorological variables. The main parameter that investigated in this study was the number of neuron of hidden layer. The performance of the prediction model was analysed and evaluated by using mean square error (MSE) and Coefficient of Determination (R2) values. With the increasing of the number of neuron of hidden layer, MSE decreased and R increased. 10 neuron of hidden layer gave the best performance among the number of neuron investigated. Due to the low performance of the prediction model, feature selection was introduced to remove irrelevant variables in data set. Random forest (RF) was grew with 200 regression trees to decide the importance of the predictors. The predictors which was less important were removed from the predictors. With the removal of the irrelevant variables, the precision of the prediction model increased with increased of the performance of the model. Besides that, the complexity of the prediction model also reduced by decreasing training time of the prediction model. The predictors removed by feature selection in this study were pressure, dew point, hourly precipitation and cumulated precipitation. Thus, it was clearly seen that the performance of prediction model with feature selection was better than prediction model without feature selection.
Contributor(s):
Sharon Ding Tiew Kui - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875007567
Language:
English
Subject Keywords:
feed-forward artificial neural network (FANN); air quality; PM2.5
First presented to the public:
6/1/2018
Original Publication Date:
7/18/2018
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Chemical Engineering
Citation:
Extents:
Number of Pages - 77
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2018-07-18 13:15:25.721
Date Last Updated
2019-01-07 11:24:32.9118
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Feature selection and model prediction of air quality using pm2.51 2018-07-18 13:15:25.721