Publisher

IJAINIJAIN

Journal International Journal of Advances in Intelligent InformaticsInternational Journal of Advances in Intelligent Informatics Abstract

This study aimed to compare the accuracy of two algorithms, hierarchical cluster and K-Means cluster, using ACFs distance for clustering stationary and non-stationary time series data. This research uses both simulation and real datasets. The simulation generates 7 stationary data models and another 7 of non-stationary data models. On the other hands, the real dataset is the daily temperature data in 34 cities in Indonesia. As a result, K-Means algorithm has the highest accuracy for both data models.

Conclusion

The experiment on simulated data and real dataset showed that the K-Means algorithm has the highest accuracy in both data models, stationary and non-stationary data.Specifically, the K-Means algorithm achieved an accuracy of 84.13286% in the simulated data and 85.Therefore, it can be concluded that K-Means is the best algorithm for classifying stationary and non-stationary time series data.

Future Research

Further research could investigate the application of deep learning techniques, such as recurrent neural networks (RNNs) or long short-term memory (LSTM) networks, to improve the accuracy of classifying stationary and non-stationary time series data, moving beyond traditional clustering approaches. Moreover, future studies could explore the integration of domain knowledge, such as meteorological factors in the case of temperature data, into the clustering process to enhance the interpretability and relevance of the results. Specifically, examining the impact of feature engineering—creating new variables from existing ones—on clustering performance could prove valuable. Additionally, investigating the robustness of these algorithms to different data preprocessing techniques, like normalization or standardization, will be helpful. Finally, research could address the challenges of scaling these methods to very large datasets by exploring the use of distributed computing frameworks or approximate clustering algorithms.

References

1. Clustering stationary and non-stationary time series based on autocorrelation distance of hierarchical... doi.org/10.26555/ijain.v3i3.98Clustering stationary and non stationary time series based on autocorrelation distance of hierarchical doi 10 26555 ijain v3i3 98

Metadata

File size	749.28 KB
Pages	7
DMCA	ReportReport