Pattern Recognition and Image Analysis

, Volume 28, Issue 4, pp 703–711

# An Exact Algorithm of Searching for the Largest Cluster in an Integer-Valued Problem of 2-Partitioning a Sequence

Mathematical Method in Pattern Recognition

## Abstract

We analyze mathematical aspects of one of the fundamental data analysis problems consisting in the search (selection) for the subset with the largest number of similar elements among a collection of objects. In particular, the problem appears in connection with the analysis of data in the form of time series (discrete signals). One of the problems in modeling this challenge is considered, namely, the problem of finding the cluster of the largest size (cardinality) in a 2-partition of a finite sequence of points in Euclidean space into two clusters (subsequences) under two constraints. The first constraint is on the choice of the indices of elements included in the clusters. This constraint simulates the set of time-admissible configurations of similar elements in the observed discrete signal. The second constraint is imposed on the value of the quadratic clustering function. This constraint simulates the level of intracluster proximity of objects. The clustering function under the second constraint is the sum (over both clusters) of the intracluster sums of squared distances between the cluster elements and its center. The center of one of the clusters is unknown and defined as the centroid (the arithmetic mean over all elements of this cluster). The center of the other cluster is the origin. Under the first constraint, the difference between any two subsequent indices of elements contained in a cluster with an unknown center is bounded above and below by some constants. It is established in the paper that the optimization problem under consideration, which models one of the simplest significant problems of data analysis, is strongly NP-hard. We propose an exact algorithm for the case of a problem with integer coordinates of its input points. If the dimension of the space is bounded by a constant, then the algorithm is pseudopolynomial.

## Keywords

time-series analysis similar elements Euclidean space sequence 2-partition longest subsequence quadratic scattering NP-hard problem integer coordinates exact algorithm fixed space dimension pseudopolynomial running time

## Preview

Unable to display preview. Download preview PDF.

## References

1. 1.
J. B. MacQueen, “Some methods for classification and analysis of multivariate observations,” in Proc. Fifth Berkeley Symposium on Mathematical Statistics and Probability (Univ. of California Press, Berkeley, 1967), Vol. 1, pp. 281–297.Google Scholar
2. 2.
M. Rao, “Cluster analysis and mathematical programming,” J. Amer. Stat. Assoc. 66, 622–626 (1971).
3. 3.
P. Hansen, B. Jaumard, and N. Mladenović, “Minimum sum of squares clustering in a low dimensional space,” J. Classif. 15 (1), 37–55 (1998).
4. 4.
P. Hansen and B. Jaumard, “Cluster analysis and mathematical programming,” Math. Program. 79 (1–3), 191–215 (1997).
5. 5.
R. A. Fisher, Statistical Methods and Scientific Inference (Hafner Press, New York, 1956).
6. 6.
D. Aloise, A. Deshpande, P. Hansen, and P. Popat, “NP–hardness of Euclidean sum–of–squares clustering,” Mach. Learn. 75 (2), 245–248 (2009).
7. 7.
P. Drineas, A. Frieze, R. Kannan, S. Vempala, V. Vinay, “Clustering large graphs via the singular value decomposition,” Mach. Learn. 56 (1–3), 9–33 (2004).
8. 8.
A. V. Dolgushev and A. V. Kel’manov, “On the algorithmic complexity of a problem in cluster analysis,” J. Appl. Ind. Math. 5 (2), 191–194 (2011).
9. 9.
M. Mahajan, P. Nimbhorkar, and K. Varadarajan, “The planar k–means problem is NP–hard,” Theor. Comput. Sci. 442, 13–21 (2012).
10. 10.
A. V. Kel’manov and S. A. Khamidullin. “An approximating polynomial algorithm for a sequence partitioning problem,” J. Appl. Ind. Math. 8 (2), 236–244 (2014).
11. 11.
A. V. Kel’manov, S. A. Khamidullin, and V. I. Khandeev, “Exact pseudopolynomial algorithm for one sequence partitioning problem,” Autom. Remote Control, 78 (1), 66–73 (2017).
12. 12.
A. V. Kel’manov, S. A. Khamidullin, and V. I. Khandeev, “A fully polynomial–time approximation scheme for a sequence 2–cluster partitioning problem,” J. Appl. Ind. Math. 10 (2), 209–219 (2016).
13. 13.
A. V. Kel’manov, S. A. Khamidullin, and V. I. Khandeev, “A randomized algorithm for a sequence 2–clustering problem,” Comput. Math. and Math. Phys. (accepted) (2018).Google Scholar
14. 14.
A. V. Kel’manov, S. A. Khamidullin, and V. I. Khandeev, “A randomized algorithm for 2–partition of a sequence,” in Analysis of Images, Social Networks and Texts, AIST 2017, Ed. by W. M. P. van der Aalst, Lecture Notes in Computer Science (Springer, Cham, 2018), Vol. 10716, pp. 297–306.Google Scholar
15. 15.
C. M. Bishop, Pattern Recognition and Machine Learning (Springer Science + Business Media, New York, 2006).
16. 16.
G. James, D. Witten, T. Hastie, and R. Tibshirani, An Introduction to Statistical Learning: With Application in R (Springer Science + Business Media, New York, 2013).
17. 17.
T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd ed. (Springer, New York, 2009).
18. 18.
C. C. Aggarwal, Data Mining: The Textbook (Springer International Publishing, Cham, 2015).
19. 19.
I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning, in Adaptive Computation and Machine Learning series (MIT Press, Cambridge, MA, 2016).
20. 20.
A. S. Shirkhorshidi, S. Aghabozorgi, T. Y. Wah, and T. Herawan, “Big data clustering: A review,” in Computational Science and Its Applications—ICCSA 2014, Ed. by B. Murgante, Lecture Notes in Computer Science (Springer, Cham, 2014), Vol. 8583, pp. 707–720.Google Scholar
21. 21.
A. K. Jain, “Data clustering: 50 years beyond–means,” Pattern Recogn. Lett. 31 (8), 651–666 (2010).Google Scholar
22. 22.
J. Pach and P. K. Agarwal, Combinatorial Geometry (Wiley, New York, 1995).
23. 23.
T.–C. Fu, “A review on time series data mining,” Eng. Appl. Artif. Intell. 24 (1), 164–181 (2011).Google Scholar
24. 24.
C. Künzer, S. Dech, and W. Wagner (Eds.), Remote Sensing Time Series, in Remote Sensing and Digital Image Processing (Springer, Cham, 2015), Vol. 22.Google Scholar
25. 25.
T. W. Liao, “Clustering of time series data–A survey,” Pattern Recogn. 38 (11), 1857–1874 (2005).
26. 26.
A. V. Kel’manov and A. V. Pyatkin, “On the complexity of a search for a subset of “similar” vectors,” Dokl. Math. 78 (1), 574–575 (2008).
27. 27.
A. V. Kel’manov and A. V. Pyatkin, “On a version of the problem of choosing a vector subset,” J. Appl. Ind. Math. 3 (4), 447–455 (2009).
28. 28.
A. V. Kel’manov and V. I. Khandeev, “A 2–approximation polynomial algorithm for a clustering problem,” J. Appl. Ind. Math. 7 (4), 515–521 (2013).
29. 29.
E. Kh. Gimadi, A. V. Kel’manov, M. A. Kel’manova, and S. A. Khamidullin, “A posteriori detection of a quasiperiodic fragment with a given number of repetitions in a numerical sequence,” Sib. Zh. Ind. Mat. 9 (1(25)), 55–74 (2006) [in Russian].Google Scholar
30. 30.
E. Kh. Gimadi, A. V. Kel’manov, M. A. Kel’manova, and S. A. Khamidullin, “A posteriori detecting a quasiperiodic fragment in a numerical sequence,” Pattern Recogn. Image Anal. 18 (1), 30–42 (2008).
31. 31.
A. E. Baburin, E. Kh. Gimadi, N. I. Glebov, and A. V. Pyatkin, “The problem of finding a subset of vectors with the maximum total weight,” J. Appl. Ind. Math. 2 (1), 32–38 (2008).
32. 32.
A. V. Kel’manov and V. I. Khandeev, “An exact pseudopolynomial algorithm for a problem of the two–cluster partitioning of a set of vectors,” J. Appl. Ind. Math. 9 (4), 497–502 (2015).
33. 33.
E. Kh. Gimadi, A. V. Pyatkin, and I. A. Rykov, “On polynomial solvability of some problems of a vector subset choice in a Euclidean space of fixed dimension,” J. Appl. Ind. Math. 4 (1), 48–53 (2010).
34. 34.
V. V. Shenmaier, “Solving some vector subset problems by Voronoi diagrams,” J. Appl. Ind. Math. 10 (4), 560–566 (2016).
35. 35.
A. V. Dolgushev and A. V. Kel’manov, “An approximation algorithm for solving a problem of cluster analysis,” J. Appl. Ind. Math. 5 (4), 551–558 (2011).
36. 36.
A. V. Dolgushev, A. V. Kel’manov, and V. V. Shenmaier, “Polynomial–time approximation scheme for a problem of partitioning a finite set into two clusters,” Proc. Steklov Inst. Math. 295 (Suppl. 1), S47–S56 (2016).Google Scholar
37. 37.
A. V. Kel’manov and V. I. Khandeev, “Fully polynomial–time approximation scheme for a special case of a quadratic Euclidean 2–clustering problem,” Comput. Math. Math. Phys. 56 (2), 334–341 (2016).
38. 38.
A. V. Kel’manov, A. V. Motkova, and V. V. Shenmaier, “An approximation scheme for a weighted two–cluster partition problem,” in Analysis of Images, Social Networks and Texts, AIST 2017, Ed. by W. M. P. van der Aalst, Lecture Notes in Computer Science (Springer, Cham, 2018), Vol. 10716, pp. 323–333.Google Scholar
39. 39.
A. V. Kel’manov and V. I. Khandeev, “A randomized algorithm for two–cluster partition of a set of vectors,” Comput. Math. Math. Phys. 55 (2), 330–339 (2015).
40. 40.
A. V. Kel’manov, V. I. Khandeev, and A. V. Panasenko, “Exact algorithms for the special cases of two hard to solve problems of searching for the largest subset,” in Proc. 7th Intern. Conf. on Analysis of Images, Social Networks, and Texts (AIST 2018), Moscow, Russia, July 5–7, 2018, Lecture Notes in Computer Science (accepted) (Springer, Cham, 2019).Google Scholar
41. 41.
A. V. Kel’manov, V. I. Khandeev, and A. V. Panasenko, “Exact algorithms for two hard to solve 2–clustering problems,” Pattern Recogn. Image Anal. (accepted) (2019).Google Scholar
42. 42.
A. V. Kel’manov and A. V. Pyatkin, “On complexity of some problems of cluster analysis of vector sequences,” J. Appl. Ind. Math. 7 (3), 363–369 (2013).
43. 43.
A. V. Kel’manov and A. V. Khamidullin, “An approximation polynomial–time algorithm for a sequence biclustering problem,” Comput. Math. Math. Phys. 55 (6), 1068–1076 (2015).
44. 44.
A. V. Kel’manov and A. V. Pyatkin, “NP–completeness of some problems of choosing a vector subset,” J. Appl. Ind. Math. 5 (3), 352–357 (2011).
45. 45.
M. R. Garey and D. S. Johnson, Computers and Intractability: A Guide to the Theory of NP–Completeness (W. H. Freeman and Co., San Francisco, 1979).
46. 46.
A. V. Kel’manov and S. A. Khamidullin, “An approximating polynomial algorithm for a sequence partitioning problem,” J. Appl. Ind. Math. 8 (2), 236–244 (2014).
47. 47.
A. V. Kel’manov and S. A. Khamidullin, “A posteriori detection of a given number of identical subsequences in a quasiperiodic sequence,” Comput. Math. Math. Phys. 41 (5), 762–774 (2001).

## Authors and Affiliations

• A. V. Kel’manov
• 1
• 2
Email author
• S. A. Khamidullin
• 1
• V. I. Khandeev
• 1
• 2
• A. V. Pyatkin
• 1
• 2
1. 1.Sobolev Institute of Mathematics, Siberian BranchRussian Academy of SciencesNovosibirskRussia
2. 2.Novosibirsk State UniversityNovosibirskRussia