Comparison of Clustering K-Means, Fuzzy C-Means, and Linkage for Nasa Active Fire Dataset

  • Institut Teknologi Adhi Tama Surabaya
  • Institut Teknologi Adhi Tama Surabaya
  • Institut Teknologi Adhi Tama Surabaya
https://doi.org/10.25139/ijair.v2i2.3030
Abstract views: 779 , PDF downloads: 298
Ethical Statement downloads: 0
Copyright Transfer Agreement downloads: 0
Hasil cek plagiasi downloads: 0
Keywords: Active fire dataset, K-Means, FCM, Linkage, Elbow Clustering.

Abstract

One of the causes of forest fires is the lack of speed of handling when a fire occurs. This can be anticipated by determining how many extinguishing units are in the center of the hot spot. To get hotspots, NASA has provided an active fire dataset. The clustering method is used to get the most optimal centroid point. The clustering methods we use are K-Means, Fuzzy C-Means (FCM), and Average Linkage. The reason for using K-means is a simple method and has been applied in various areas. FCM is a partition-based clustering algorithm which is a development of the K-means method. The hierarchical based clustering method is represented by the Average Linkage method.  The measurement technique that uses is the sum of the internal distance of each cluster. Elbow evaluation is used to evaluate the optimal cluster. The results obtained after conducting the K-Means trial obtained the best results with a total distance of 145.35 km, and the best clusters from this method were 4 clusters. Meanwhile, the total distance values obtained from the FCM and Linkage methods were 154.13 km and 266.61 km.



Downloads

Download data is not yet available.

References

[1] R. R. Muhima, M. Kurniawan, and O. T. Pambudi, “A LOF K - Means Clustering on Hotspot Data,” Int. J. Artif. Intell. Robot., vol. 2, no. 1, pp. 29–33, 2020, doi: 10.25139/ijair.v2i1.2634.
[2] E. Çolak and F. Sunar, “The importance of ground-truth and crowdsourcing data for the statistical and spatial analyses of the NASA FIRMS active fires in the Mediterranean Turkish forests,” Remote Sens. Appl. Soc. Environ., vol. 19, no. April, p. 100327, 2020, doi: 10.1016/j.rsase.2020.100327.
[3] T. V Loboda, L. Giglio, L. Boschetti, and C. O. Justice, “Regional fire monitoring and characterization using global NASA MODIS fire products in dry lands of Central Asia,” Front. Earth Sci., vol. 6, no. 2, pp. 196–205, 2012, doi: 10.1007/s11707-012-0313-3.
[4] R. V. Virgil Petrescu, R. Aversa, T. M. Abu-Lebdeh, A. Apicella, and F. I. T. Petrescu, “NASA Satellites Help us to Quickly Detect Forest Fires,” Am. J. Eng. Appl. Sci., vol. 11, no. 1, pp. 288–296, 2018, doi: 10.3844/ajeassp.2018.288.296.
[5] P. Li, C. Xiao, Z. Feng, W. Li, and X. Zhang, “Occurrence frequencies and regional variations in Visible Infrared Imaging Radiometer Suite (VIIRS) global active fires,” Glob. Chang. Biol., vol. 26, no. 5, pp. 2970–2987, 2020, doi: 10.1111/gcb.15034.
[6] X. Wei, G. Wang, T. Chen, D. F. T. Hagan, and W. Ullah, “A spatio-temporal analysis of active fires over China during 2003-2016,” Remote Sens., vol. 12, no. 11, 2020, doi: 10.3390/rs12111787.
[7] A. A. Pereira et al., “Burned area mapping in the Brazilian Savanna using a one-class support vector machine trained by active fires,” Remote Sens., vol. 9, no. 11, 2017, doi: 10.3390/rs9111161.
[8] O. C. D. Anejionu, G. A. Blackburn, and J. D. Whyatt, “Detecting gas flares and estimating flaring volumes at individual flow stations using MODIS data,” Remote Sens. Environ., vol. 158, pp. 81–94, 2015, doi: 10.1016/j.rse.2014.11.018.
[9] D. Hand, H. Mannila, and P. Smyth, Principles of Data Mining, vol. 2001. 2001.
[10] W. M. J. Mohammed J. Zaki, Data Mining and Analysis: Fundamental Concepts and Algorithms, vol. 27. Cambridge University, 2013.
[11] A. Martino, A. Rizzi, and F. M. F. Mascioli, “Distance Matrix Pre-Caching and Distributed Computation of Internal Validation Indices in k-medoids Clustering,” Proc. Int. Jt. Conf. Neural Networks, vol. 2018-July, pp. 1–8, 2018, doi: 10.1109/IJCNN.2018.8489101.
[12] P. A. Murena, J. Sublime, B. Matei, and A. Cornuéjols, “An information theory based approach to multisource clustering,” IJCAI Int. Jt. Conf. Artif. Intell., vol. 2018-July, no. July, pp. 2581–2587, 2018, doi: 10.24963/ijcai.2018/358.
[13] V. S. Akondi, V. Menon, J. Baudry, and J. Whittle, “Novel K-Means Clustering-based Undersampling and Feature Selection for Drug Discovery Applications,” in 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2019, pp. 2771–2778, doi: 10.1109/BIBM47256.2019.8983213.
[14] M. Misuraca, M. Spano, and S. Balbi, “BMS: An improved Dunn index for Document Clustering validation,” Commun. Stat. - Theory Methods, vol. 48, no. 20, pp. 5036–5049, 2019, doi: 10.1080/03610926.2018.1504968.
[15] J. Lipor and L. Balzano, “Clustering quality metrics for subspace clustering,” Pattern Recognit., vol. 104, p. 107328, 2020, doi: 10.1016/j.patcog.2020.107328.