Fluctuated Fitting Under the \(\ell _1\)-metric

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10336)


We consider the problem of fitting a given sequence of integers by an \((\alpha ,\beta )\)-fluctuated one. For a sequence of numbers, those elements which are larger than their direct precursors are called ascends, those elements which are smaller than their direct precursors are called descends. A sequence is said to be \((\alpha ,\beta )\)-fluctuated if there is a descend between any \(\alpha +1\) ascends and an ascend between any \(\beta +1\) descends; or equivalently, if it has at most \(\alpha \) consecutive ascends and at most \(\beta \) consecutive descends, when adjacent equal values are ignored.

Given a sequence of integers \(\mathbf {a}=(a_1,\ldots ,a_n)\) and two parameters \(\alpha ,\beta \) in [1, n], we compute (1) a sequence \(\mathbf {b}=(b_1,\ldots ,b_n)\) of integers that is \((\alpha ,\beta )\)-fluctuated and is closest to \(\mathbf {a}\) among all such sequences; (2) a sequence \(\mathbf {b}'=(b'_1,\ldots ,b'_n)\) of integers that is \((\alpha ,\beta )\)-fluctuated and is bounded by \(\mathbf {a}\) (i.e. \(b'_i\le a_i\) for all i) and is closest to \(\mathbf {a}\) among all such sequences. We measure the distance between sequences under \(\ell _1\) metric.

Our algorithm runs in \(O((\alpha +\beta )\cdot n)\) time, which is linear when \(\alpha ,\beta \) are considered as constants. We also show that a variation of our problem can be solved in the same time complexity. We achieve our result mainly by exploiting and utilizing the property of the closest sequence.


Curve approximation Histogram Fitting 


  1. 1.
    Aronov, B., Asano, T., Katoh, N., Mehlhorn, K., Tokuyama, T.: Polyline fitting of planar points under min-sum criteria. Int. J. Comput. Geom. Appl. 16, 97–116 (2006)MathSciNetCrossRefzbMATHGoogle Scholar
  2. 2.
    Chen, D.Z., Wang, H.: Approximating points by a piecewise linear function. Algorithmica 66(3), 682–713 (2013)MathSciNetCrossRefzbMATHGoogle Scholar
  3. 3.
    Chen, D.Z., Healy, M.A., Wang, C., Xu, B.: Geometric algorithms for the constrained 1-D K-means clustering problems and IMRT applications. In: Preparata, F.P., Fang, Q. (eds.) FAW 2007. LNCS, vol. 4613, pp. 1–13. Springer, Heidelberg (2007). doi: 10.1007/978-3-540-73814-5_1 CrossRefGoogle Scholar
  4. 4.
    Chen, D., Wang, C., Wang, H.: Representing a functional curve by curves with fewer peaks. Discret. Comput. Geom. 46(2), 334–360 (2011)MathSciNetCrossRefzbMATHGoogle Scholar
  5. 5.
    Chun, J., Sadakane, K., Tokuyama, T.: Linear time algorithm for approximating a curve by a single-peaked curve. Algorithmica 44(2), 103–115 (2006)MathSciNetCrossRefzbMATHGoogle Scholar
  6. 6.
    Chun, J., Sadakane, K., Tokuyama, T., Yuki, M.: Peak-reducing fitting of a curve under the \(l_p\) metric. Interdisc. Inf. Sci. 2, 191–197 (2005)MathSciNetGoogle Scholar
  7. 7.
    Cleju, I., Fränti, P., Wu, X.: Clustering based on principal curve. In: Kalviainen, H., Parkkinen, J., Kaarna, A. (eds.) SCIA 2005. LNCS, vol. 3540, pp. 872–881. Springer, Heidelberg (2005). doi: 10.1007/11499145_88 CrossRefGoogle Scholar
  8. 8.
    Douglas, D.H., Peucker, T.K.: Algorithms for the Reduction of the Number of Points Required to Represent a Digitized Line or its Caricature. Wiley, Hoboken (2011). pp. 15–28CrossRefGoogle Scholar
  9. 9.
    Fournier, H., Vigneron, A.: A deterministic algorithm for fitting a step function to a weighted point-set. Inf. Process. Lett. 113(3), 51–54 (2013)MathSciNetCrossRefzbMATHGoogle Scholar
  10. 10.
    Fournier, H., Vigneron, A.: Fitting a step function to a point set. In: Halperin, D., Mehlhorn, K. (eds.) ESA 2008. LNCS, vol. 5193, pp. 442–453. Springer, Heidelberg (2008). doi: 10.1007/978-3-540-87744-8_37 CrossRefGoogle Scholar
  11. 11.
    Goodrich, M.: Efficient piecewise-linear function approximation using the uniform metric. Discret. Comput. Geom. 14(1), 445–462 (1995)MathSciNetCrossRefzbMATHGoogle Scholar
  12. 12.
    Guha, S., Shim, K.: A note on linear time algorithms for maximum error histograms. Knowl. Data Eng. 19(7), 993–997 (2007)CrossRefGoogle Scholar
  13. 13.
    Guha, S., Koudas, N., Shim, K.: Data-streams and histograms. In: Proceedings of 33rd Symposium on Theory of Computing, STOC 2001, pp. 471–475. ACM (2001)Google Scholar
  14. 14.
    Haiminen, N., Gionis, A., Laasonen, K.: Algorithms for unimodal segmentation with applications to unimodality detection. Knowl. Inf. Syst. 14(1), 39–57 (2008)CrossRefGoogle Scholar
  15. 15.
    Luebke, D.P.: A developer’s survey of polygonal simplification algorithms. IEEE Comput. Graph. Appl. 21(3), 24–35 (2001)CrossRefGoogle Scholar
  16. 16.
    Ramesh, N., Yoo, J.H., Sethi, I.: Thresholding based on histogram approximation. IEEE Proc. Vis. Image Sig. Process. 142(5), 271–279 (1995)CrossRefGoogle Scholar
  17. 17.
    Stout, Q.F.: Unimodal regression via prefix isotonic regression. Comput. Stat. Data Anal. 53(2), 289–297 (2008)MathSciNetCrossRefzbMATHGoogle Scholar
  18. 18.
    Stout, Q.F.: Isotonic regression via partitioning. Algorithmica 66(1), 93–112 (2013)MathSciNetCrossRefzbMATHGoogle Scholar
  19. 19.
    Stout, Q.F.: Isotonic regression for multiple independent variables. Algorithmica 71(2), 450–470 (2015)MathSciNetCrossRefzbMATHGoogle Scholar
  20. 20.
    Tokuyama, T.: Recent progress on geometric algorithms for approximating functions: toward applications to data analysis. Electron. Commun. Jpn. (Part III: Fundam. Electron. Sci.) 90(3), 1–12 (2007)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.University of Hong KongPokfulamHong Kong SAR, China

Personalised recommendations