The Efficiency of Measure Particulate Matter Device using Classification Techniques and Association Rules Discovery for Factor Analysis Influencing Particulate Matter

Pongpat Singsri; Sahapat Chalachai

doi:10.14456/nujst.2021.27

Authors

Pongpat Singsri Information and Communication Technology, Science and Technology, Rajamangala University of Technology Tawan-ok, Chonburi, 20110, Thailand
Sahapat Chalachai Agricultural Engineering and Technology, Agriculture and Natural Resources, Rajamangala University of Technology Tawan-ok, Chonburi, 20110, Thailand

DOI:

https://doi.org/10.14456/nujst.2021.27

Keywords:

Particulate Matter 2.5, microcontroller, Decision Tree Algorithm, Apriori Algorithm

Abstract

The objectives of the research were to develop low-cost Measure Particulate Matter Device (PM_2.5) with real-time display and analyze data using classification techniques and association rules discovery. The developed device has a microcontroller with a PM_2.5 dust detector using infrared light, temperature and humidity sensor, and raindrop sensor. The device is being installed at the weather station in Laem Chabang Municipal Stadium (32T), Chonburi, Thailand. The weather station belongs to the Air Quality and Noise Management Bureau, Pollution Control Department, Ministry of Natural Resources and Environment. The data is collected every 20 seconds for 50 days, received a total of 196,023 records. The collected data is divided into 2 data sets; the first data set is the data that device and weather station measures at the same time and the second data set are the average hourly data of that device and weather station measure. The two sets of data compared to measure the efficiency of the device. The data was analyzed by data mining using Classification Techniques with Decision Tree algorithm and Association Rule with the Apriori algorithm. The results of this research showed that the device has the accuracy in measuring at ±0.5 (±5.77%) with hourly data set and has the accuracy in measuring at ±1.0 (±8.46%) with the hourly average data set. Decision Tree algorithm has accuracy for the forecast at 62.36%. Apriori algorithm gave the highest confidence in Association Rules at 78% with PM_2.5 between 20.00 – 24.99 µg/m³ relation to a temperature between 30.00^o – 34.99^o C

References

AirVisual. (2020). Air Quality Index (AQI). Retrieved from https://www.iqair.com

Ataei, S. H., Mohammadzadeh, A., & Abkar, A. A. (2015). Using Decision Tree Method for Dust Detection from MODIS Satellite Image. Journal of Geomatics Science and Technology, 4(4), 151-160.

Chiang Mai Province Offices for Natural Resources and Environment. (2019). The air quality is at a level that exceeds the standard value. Retrieved from https://www.chiangmai.mnre.go.th/th/news/more/1380

Dailynews. (2015). Air pollution in in Beijing soared 20 times. Retrieved from https://www.dailynews.co.th/foreign/294174

Intra, P. (2009). Real-Time Measuring and Sampling of PM2.5 and PM10. Research promotion and development section, Research management mission, National Research council of Thailand. Retrieved from https://www.priv.nrct.go.th/shopping/home/show_product.php?research_id=517

Manager Online. (2013). Installed Measure Particulate Matter Device (Real Time) at Lampang Child Center. Retrieved from https://www.manager.co.th/Local/ViewNews.aspx?NewsID=9560000043806

Rattanapoka, C. (2013). Teaching documents 030523111 Introduction to Artificial Intelligence. Bangkok: King Mongkut's University of Technology North Bangkok.

Real-time Air Quality Index.(2020). NRCT, BKK Air Pollution. Retrieved from https://www.aqicn.org/city/thailand/bkk/nrct/

Research Institute for Health Science. (2012). Research Institute for Health Science are successful invention of the small dust particle analyzer to stimulate the resolution of forest fire smog from the local level, General Division, Office of the University, Chiang Mai University. Retrieved from http://www.prcmu.cmu.ac.th/perin_detail.php?perin_id=327

Singsri, P., Phiromlap, S., Saramas, S., & Koodsamrong, R. (2018). Factor Analysis Influencing Student Retire Using Classification Technique Case Study Rajamangala University of Technology Tawan-ok, Bangphra Campus. LRU Research Conference 2018, 363 – 371.

Singsri, P., Suksawatchon, J., & Suksawatchon, U. (2013). Apply of Classification Techniques for Factor Analysis Influencing Quail Gender Identification Considering from the External Factors of Quail Eggs. National Conference on Computing and Information Technology, 9, 240 – 247.

Songsiri, C., Rakthanmanon, T., & Waiyamai, K. (2001). Applying a data mining technique to help students in selecting their majors. Kasetsart University Annual Conference: Engineering, 39, 43-50.

Suksawatchon, J., Suksawatchon, U., & Singsri, P. (2014). Feature Selection and Efficiency Comparison of Classification Techniques for Quail Gender Forecasting from the External Factors of Quail Eggs, National Conference on Computing and Information Technology, 10, 515 – 521.

Suksawatchon, U., & Singsri, P. (2014). Factor Analysis Influencing Quail Gender Classification Considering from the External Factors of Quail Eggs Using Classification Techniques. International Joint Conference on Computer Science and Software Engineering, 11, 297-301.

Supasri, T., Intra, P., Jomjunyong, S., & Sampattagul, S. (2018). Evaluation of Particulate Matter Concentration by Using a Wireless Sensor System for Continuous Monitoring of Particulate Air Pollution in Northern of Thailand. Journal of Innovative Technology Research, 2(1), 69–83.