Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
184 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Memetic Differential Evolution Methods for Semi-Supervised Clustering (2403.04322v2)

Published 7 Mar 2024 in math.OC, cs.LG, and cs.NE

Abstract: In this paper, we propose an extension for semi-supervised Minimum Sum-of-Squares Clustering (MSSC) problems of MDEClust, a memetic framework based on the Differential Evolution paradigm for unsupervised clustering. In semi-supervised MSSC, background knowledge is available in the form of (instance-level) "must-link" and "cannot-link" constraints, each of which indicating if two dataset points should be associated to the same or to a different cluster, respectively. The presence of such constraints makes the problem at least as hard as its unsupervised version and, as a consequence, some framework operations need to be carefully designed to handle this additional complexity: for instance, it is no more true that each point is associated to its nearest cluster center. As far as we know, our new framework, called S-MDEClust, represents the first memetic methodology designed to generate a (hopefully) optimal feasible solution for semi-supervised MSSC problems. Results of thorough computational experiments on a set of well-known as well as synthetic datasets show the effectiveness and efficiency of our proposal.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (41)
  1. doi:10.1007/BF02614317.
  2. doi:10.1007/s10994-009-5103-0.
  3. doi:10.2307/2346830.
  4. doi:10.1109/TIT.1982.1056489.
  5. doi:10.1016/S0031-3203(02)00060-2.
  6. doi:10.1016/0031-3203(91)90097-O.
  7. doi:10.1016/0031-3203(95)00022-R.
  8. doi:10.1016/S0031-3203(99)00137-5.
  9. doi:10.1016/S0167-8655(97)00122-0.
  10. doi:10.1007/s10898-014-0171-5.
  11. doi:10.1016/j.patcog.2015.11.011.
  12. doi:10.1016/j.patcog.2018.12.022.
  13. doi:10.1016/j.patcog.2021.107849.
  14. doi:10.1023/A:1008202821328.
  15. doi:10.1137/1.9781611972788.3.
  16. doi:10.1016/j.cor.2021.105299.
  17. doi:10.1145/1148170.1148242.
  18. doi:10.1016/j.ipm.2008.03.001.
  19. doi:10.1007/978-3-030-61527-7_4.
  20. doi:10.1109/IEEM45057.2020.9309775.
  21. doi:10.1137/1.9781611972740.31.
  22. doi:10.1137/1.9781611972757.13.
  23. doi:10.1137/1.9781611974348.33.
  24. doi:10.1016/j.cor.2020.104979.
  25. doi:10.48550/arXiv.2212.14437.
  26. doi:10.1007/s10618-008-0104-3.
  27. doi:10.1007/s10107-010-0349-7.
  28. doi:10.1016/j.artint.2015.05.006.
  29. doi:10.3233/978-1-61499-672-9-462.
  30. doi:10.1016/j.cor.2022.105958.
  31. doi:10.1016/j.ins.2022.05.035.
  32. doi:10.1007/s10898-021-01047-6.
  33. doi:10.1145/1273496.1273522.
  34. doi:10.1016/j.cor.2013.09.010.
  35. doi:10.1002/nav.3800020109.
  36. doi:10.1142/9789814324700_0104.
  37. doi:10.1007/978-3-642-24425-4_30.
  38. doi:10.1007/BF01908075.
  39. doi:10.1109/TSMC.1987.4309069.
  40. doi:10.1016/j.eswa.2023.121953.
  41. doi:10.1007/s101070100263.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now