Perbandingan Complexity Invariant Distance (CID) dan Dynamic Time Warping (DTW) dalam Analisis Klaster Deret Waktu pada Nilai Tukar Petani di Indonesia

Penulis

  • Laila Fathiyaturrahmi Institut Teknologi Sepuluh Nopember, Surabaya
  • Andriano Institut Teknologi Sepuluh Nopember, Surabaya
  • Harista Almiatus Soleha Institut Teknologi Sepuluh Nopember, Surabaya
  • Dedy Dwi Prastyo Institut Teknologi Sepuluh Nopember, Surabaya

DOI:

https://doi.org/10.25126/jtiik.124

Kata Kunci:

Cluster time series, complexity invariant distance, dynamic time warping, farmer exchange rate

Abstrak

Analisis klaster yang merupakan bagian dari data mining yang membagi data kedalam beberapa kelompok berdasarkan kedekatan karakteristik tertentu. Konsep utama dalam klaster adalah memaksimalkan kedekatan data di dalam klaster dan meminimalkan kesamaan data antar klaster. Analisis klaster juga bisa digunakan pada berbagai jenis data termasuk data deret waktu.  Pengukuran kesamaan menjadi hal yang utama pada analisis klaster. Metode yang bisa digunakan dalam pengukuran jarak yaitu Complexity Invariant Distance (CID) dan Dynamic Time Warping (DTW). Analisis pengukuran jarak CID dan DTW dapat digunakan pada pengelompokkan data deret waktu salah satunya pada data Nilai Tukar Petani (NTP). NTP dapat menggambarkan daya beli petani karena diperoleh dari perbandingan indeks harga yang diterima petani dibandingkan dengan yang harus dibayarnya, atau dapat dinyatakan sebagai kemampuan petani dalam memnuhi kebutuhan sehari-hari dari hasil pertanian. Sehingga dilakukan analisis untuk membandingkan metode pengukuran jarak CID dan DTW pada klastering data deret waktu pada nilai tukar petani pada 34 Provinsi di Indonesia. Hasil analisis yang diakukan menunjukkan klaster terbaik adalah pengklasteran dengan banyak klaster dua (k=2) menggunakan ukuran jarak CID terlihat dari nilai silhouette 0.8776 yang lebih tinggi dibandingkan klaster lain. Dimana klaster satu terdiri dari 25 Provinsi dan klaster dua terdiri dari 9 Provinsi.

 

Abstract

Cluster analysis is a part of data mining which divides data into several groups based on the proximity of certain characteristics. The main concept in clusters is to maximize data similarity within clusters and minimize data similarity between clusters. Cluster analysis can also be used on various types of data, including time series data. Measuring similarity is the main thing in cluster analysis. The methods that can be used to measure distance are Complexity Invariant Distance (CID) and Dynamic Time Warping (DTW). CID and DTW distance measurement analysis can be used to group time series data, one of which is Farmer’s Terms of Trade (NTP) data. The farmer's terms of trade is a ratio between the price index received by farmers and the price index paid by farmers. In general, it can be interpreted as the farmer's ability to meet their daily needs through agricultural products. So an analysis was carried out to compare the CID and DTW distance measurement methods in clustering time series data on farmer’s terms of trade according to 34 provinces in Indonesia. The results of this analysis show that the best cluster is clustering with two clusters (k=2) using the CID distance measure because it has the highest silhouette coefficient value, namely 0.8776. Where cluster one consists of 25 provinces and cluster two consists of 9 provinces.

Downloads

Download data is not yet available.

Referensi

ABBASIMEHR, H., PAKI, R. AND BAHRINI, A., 2022. A novel approach based on combining deep learning models with statistical methods for COVID-19 time series forecasting. Neural Computing and Applications, 4, pp.3135–3149.

ADNYANI, P.W. AND SIHOMBING, L.R.P., 2021. Analisis Cluster Time Series Dalam Pengelompokan Provinsi Di Indonesia Berdasarkan Nilai PDRB. Lppm Bina Bangsa, [online] 1(1), pp.47–54.

AGHABOZORGI, S., SEYED SHIRKHORSHIDI, A. AND YING WAH, T., 2015. Time-series clustering - A decade review. Information Systems, [online] 53, pp.16–38. https://doi.org/10.1016/j.is.2015.04.007.

BATISTA, G.E.A.P.A., WANG, X. AND KEOGH, E.J., 2011. A complexity-invariant distance measure for time series. Proceedings of the 11th SIAM International Conference on Data Mining, SDM 2011, pp.699–710. https://doi.org/10.1137/1.9781611972818.60.

DING, H., TRAJCEVSKI, G., SCHEUERMANN, P., WANG, X. AND KEOGH, E., 2008. Querying and mining of time series data. Proceedings of the VLDB Endowment, 1(2), pp.1542–1552. https://doi.org/10.14778/1454159.1454226.

HAIR, J.F., BLACK, W.C. AND BABIN, B.J., 2010. Multivariate Data Analysis.

JOHNSON, R.A., and WICHERN, D.W., 2007. Applied Multivariate Statistical Analysis. US Patent 3,123,114.

KAUFMAN, L. AND ROUSSEEUW, P.J., 1990. Finding Groups in Data: An Introduction to Cluster Analysis. pp.1–67. https://doi.org/10.1002/9780470316801.ch1.

KAUFMAN, L. AND ROUSSEEUW, P.J., 2006. Finding Groups in Data: An introduction to Cluster ANalysis. Wiley-interscience. New Jersey: John Wiley & Sons.

MUNTHE, A.D., 2019. Penerapan clustering time series untuk menggerombolkan provinsi di Indonesia berdasarkan nilai produksi padi. urnal Litbang Sukowati: Media Penelitian dan Pengembangan, 2.

PAPARRIZOS, J. AND GRAVANO, L., 2017. Fast and accurate time-series clustering. ACM Transactions on Database Systems, 42(2). https://doi.org/10.1145/3044711.

RIYADI, M.A.A., PRATIWI, D.S., IRAWAN, A.R. AND FITHRIASARI, K., 2017. Clustering stationary and non-stationary time series based on autocorrelation distance of hierarchical and K-means algorithms. International Journal of Advances in Intelligent Informatics, 3(3), pp.154–160. https://doi.org/10.26555/ijain.v3i3.98.

SAKOE, H. AND CHIBA, S., 1978. Dynamic programming algorithm optimization for spoken word recognition. IEEE Transactions on Acoustics, Speech, and Signal Processing, 26(1), pp.43–49. https://doi.org/10.1109/TASSP.1978.1163055.

SHANG, D., SHANG, P. AND LIU, L., 2019. Multidimensional scaling method for complex time series feature classification based on generalized complexity-invariant distance. Nonlinear Dynamics, [online] 95(4), pp.2875–2892. https://doi.org/10.1007/s11071-018-4728-6.

WARREN LIAO, T., 2005a. Clustering of time series data - A survey. Pattern Recognition, 38(11), pp.1857–1874. https://doi.org/10.1016/j.patcog.2005.01.025.

Diterbitkan

29-08-2025

Terbitan

Bagian

Ilmu Komputer

Cara Mengutip

Perbandingan Complexity Invariant Distance (CID) dan Dynamic Time Warping (DTW) dalam Analisis Klaster Deret Waktu pada Nilai Tukar Petani di Indonesia. (2025). Jurnal Teknologi Informasi Dan Ilmu Komputer, 12(4), 771-778. https://doi.org/10.25126/jtiik.124