Main Article Content

Authors

Archana Rajesh Meshram

Dr. Peer Javaid Ahmad

Abstract

One effective iterative method for estimating parameters in latent variable probabilistic models is the Expectation-Maximization (EM) algorithm, which finds extensive use in classification issues. A versatile probabilistic framework for modeling data from multiple Gaussian distributions, Gaussian Mixture Models (GMMs) employ EM for parameter estimation, which is explored in this article. Problems with class membership estimation (including means, covariances, and mixing coefficients) are the primary focus of this study. Starting with initial parameter estimations, estimating posterior probabilities in the E-step, and repeatedly updating model parameters in the M-step until convergence, the technique entails applying the EM algorithm on both synthetic and real-world datasets. In order to measure performance, we compare it to K-means and other baseline classifiers using measures like classification accuracy, Adjusted Rand Index (ARI), and log-likelihood progression. When data is distributed according to a multimodal Gaussian distribution, the findings show that EM-GMM provides better parameter estimation and better classification accuracy. Initialization procedures have a substantial effect on performance, as convergence study further shows. In addition to reiterating the EM algorithm's usefulness for probabilistic classification, this paper also describes its advantages and disadvantages and offers suggestions for improving its performance in settings with high-dimensional data and noise.

Downloads

Download data is not yet available.

Article Details

Section

Articles

References

  1. McLachlan, G., & Peel, D. (2023). Finite Mixture Models. Wiley Series in Probability and Statistics. Wiley.
  2. Bishop, C. M. (2022). Pattern Recognition and Machine Learning. Springer.
  3. Dempster, A. P., Laird, N. M., & Rubin, D. B. (2021). Maximum Likelihood from Incomplete Data via the EM Algorithm. Journal of the Royal Statistical Society: Series B (Methodological), 39(1), 1-38.
  4. McLachlan, G., & Krishnan, T. (2021). The EM Algorithm and Extensions. Wiley-Interscience.
  5. Xu, L., Jordan, M. I., & Hinton, G. E. (2022). An Alternative View of the EM Algorithm for Gaussian Mixture Models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 41(3), 665-679.
  6. Zhou, Z., & Feng, J. (2023). Robust EM Algorithm for Gaussian Mixture Models in High-Dimensional Data. Neurocomputing, 525, 41-52.
  7. Likas, A., & Galatsanos, N. P. (2022). A Variational EM Algorithm for Classification with Gaussian Mixture Models. Pattern Recognition Letters, 142, 122-130.
  8. Tang, X., & Sutherland, A. (2023). Efficient Parameter Estimation for GMMs using Accelerated EM. Information Sciences, 610, 428-440.
  9. Zhang, Q., & Wang, J. (2022). Adaptive EM Algorithm for Gaussian Mixture Model-based Image Classification. IEEE Access, 10, 35780-35791.
  10. Guan, J., & Chen, L. (2023). Semi-supervised Classification Using EM Algorithm in Gaussian Mixture Models. Expert Systems with Applications, 213, 119042.
  11. Kim, D., & Park, H. (2021). EM Algorithm with Penalized Likelihood for GMM Clustering. Computational Statistics & Data Analysis, 154, 107050.
  12. Li, H., & Shen, H. (2022). Online EM Algorithm for Dynamic Gaussian Mixture Models in Streaming Data Classification. IEEE Transactions on Neural Networks and Learning Systems, 33(4), 1563-1574.
  13. Ahmed, M., & Lee, S. (2023). Improved EM Algorithm for Gaussian Mixture Models in Speech Recognition Systems. Applied Soft Computing, 126, 109453.
  14. Wang, X., & Zhang, Y. (2022). EM-Based Parameter Estimation for Gaussian Mixture Models with Missing Data. Pattern Recognition, 124, 108468.
  15. Chen, T., & Zhao, J. (2023). A Fast Converging EM Algorithm for Gaussian Mixture Model Clustering. Journal of Computational and Graphical Statistics, 32(2), 456-470.