Pengembangan Model Fast Incremental Gaussian Mixture Network (IGMN) pada Interpolasi Spasial

Authors

  • Prati Hutari Gani Telkom University, Bandung
  • Gusti Ayu Putri Saptawati Institut Teknologi Bandung, Bandung

DOI:

https://doi.org/10.30865/mib.v6i1.3490

Keywords:

Fast Incremental Gaussian Mixture Network Model, Incremental Gaussian Mixture Network, Spatial Interpolation, Variogram, Machine Learning

Abstract

Gathering geospatial information in an organization is one of the most critical processes to support decision-making and business sustainability. However, many obstacles can hinder this process, like uncertain natural conditions and a large geographical area. This problem causes the organization only to obtain a few sample points of observation, resulting in incomplete information. The data incompleteness problem can be solved by applying spatial interpolation to estimate or determine the value of unavailable data. Spatial interpolation generally uses geostatistical methods. These geostatistical methods require a variogram as a model built based on the knowledge and input of geostatistic experts. The existence of this variogram becomes a necessity to implement these methods. However, it becomes less suitable to be applied to organizations that do not have geostatistics experts. This research will develop a Fast IGMN model in solving spatial interpolation. In this study, results of the modified Fast IGMN model in spatial interpolation increase the interpolation accuracy. Fast IGMN without modification produces MSE = 1.234429691, while using Modified Fast IGMN produces MSE = 0.687391. The MSE value of the Fast IGMN-Modification model is smaller, which means that the smaller the MSE value, the higher the accuracy of the interpolation results. This modified Fast IGMN model can solve problems in gathering information for an organization that does not have geostatistics experts in the spatial data modeling process. However, it needs to be developed again with more varied input data.

References

I. A. Alonso, J. C. Verdún and E. T. Caro, "Description of the Structure of the IT Demand Management Process Framework," International Journal of Information Management, vol. 37, p. 1461–1473, 2017.

S. Bhattacharjee, P. Mitra and S. K. Ghosh, "Spatial Interpolation to Predict Missing Attributes in GIS Using Semantic Kriging," 2014.

G. Vanhoy, H. Volos, C. E. C. Bastidas and T. Bose, "A Spatial Interpolation Method for Radio Frequency Maps based on the Discrete Cosine Transform," in IEEE Military Communications Conference, 2013.

M. F. Yazar, E. Ozelkan and B. B. Üstündağ, "Multi-parameter Spatial Interpolation of Solar Radiation in Heterogeneous Structured Agricultural Areas," in The Third International Conference on Agro-Geoinformatics, 2014.

A. Nurhadiyatna, A. Sunaryani, Y. Sudriani and A. L. Latifah, "2D Spatial Interpolation for Water Quality Parameter Distribution in Maninjau Lake," in International Conference on Computer, Control, Informatics and its Applications, Jakarta, 2016.

C. Jing, M. Du, P. Dai, H. Wei and H. Liu, "Research on accuracy assessment of urban rainfall spatial interpolation from gauges data," in IEEE Geoscience and Remote Sensing Symposium, 2014.

P. Tiengrod and W. Wongseree, "A Comparison of Spatial Interpolation Methods for Surface Temperature in Thailand," in International Computer Science and Engineering Conference, 2013.

E. Oktavia, Widyawan and I. W. Mustika, "Inverse Distance Weighting and Kriging Spatial Interpolation for Data Center Thermal Monitoring," in 1st International Conference on Information Technology, Information Systems and Electrical Engineering (ICITISEE), Yogyakarta, 2016.

M. R. Heinen, P. M. Engel and R. C. Pinto, " IGMN: An Incremental Gaussian Mixture Network that Learns Instantaneously from Data Flows," in Proc VIII Encontro Nacional de Inteligencia Artificial, 2011.

S. A. F. Soares, G. S. Neto and M. Roisenberg, "Improving the Incremental Gaussian Mixture Neural Network Model for Spatial Interpolation and Geostatistical Simulation," in International Joint Conference on Neural Networks (IJCNN), 2016.

R. Pinto and P. Engel, "A fast incremental gaussian mixture model," PLos ONE, p. 10, 2015.

R. C. Pinto and P. M. Engel, "Scalable and Incremental Learning of Gaussian Mixture Models," 2017.

Downloads

Published

2022-01-25

Issue

Vol. 6 No. 1 (2022): Januari 2022

Section

Articles

License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License

Authors who publish with this journal agree to the following terms:

  1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under Creative Commons Attribution 4.0 International License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
  2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
  3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (Refer to The Effect of Open Access).