Published

2020-03-02

Preserving data moments in density estimation via diffusion using the finite element method

Preservación de momentos de datos en estimación de densidades vía difusión usando el método de elementos finitos

Keywords:

KDE, Diffusion equation, Moments preserving evolution, FEM, Lagrange Multipliers (en)
KDE, Ecuación de Difusión, evolución de preservación de momentos, FEM, Multiplicadores de Lagrange (es)

Authors

  • Keith Y. Patarroyo Université de Montréal
  • Juan Galvis Universidad Nacional de Colombia
  • Francisco Gómez Universidad Nacional de Colombia
We design a two-dimensional density estimation scheme via diffusion that conserves the first order moments and the total mass in the estimation process. In order to conserve the first order moments and the total mass throughout the time iteration, a non-local boundary condition is imposed to the diffusion operator. A discrete method is realized by using the finite element method where the boundary condition is weakly imposed using Lagrange multipliers that leads to the solution of a saddle point problem. We show some numerical examples in different geometries using FeniCS.
Se diseña un esquema de estimación de densidades vía difusión que conserva los momentos de primer orden y la masa total en el proceso de estimación. Para poder conservar los momentos de primer orden y la masa total a través del tiempo, se impone una condición de frontera no local al operador de difusión. Un método discreto es propuesto usando el método de elementos finitos donde las condiciones de frontera son impuestas débilmente usando multiplicadores de Lagrange que llevan a un problema de punto de silla. Mostramos algunos experimentos numéricos con distintas geometrías usando FEniCS.

References

13. Poisson equation with pure Neumann boundary conditions, The FEniCS Project, https://fenicsproject.org/olddocs/dolfin/2016.2.0/python/demo/documented/neumann-poisson/python/documentation.html.

2D Density Estimation via Diffusion with FEM-Neumann Boundary Conditions. Patarroyo K., Available online, https://goo.gl/Qlpruk.

C. Geuzaine and J. -F. Remacle, Gmsh (C), (1997-2017). https://gmsh.info/.

Rocket Clipart #18923, WikiClipArt, https://wikiclipart.com/rocket-clipart_18923/.

Cadcorp Spatial Information System R (Cadcorp SIS R), [A spatial statistics program for the analysis of crime incident locations], 2017, https://www.icpsr.umich.edu/CrimeStat/ CrimeStat R.

National Institute of Justice, Hot Spot Mapping Using KDE, 2017, https://www.cadcorp.com/about-us/.

M. S. Alnaes, J. Hake J. Blechta, B. Kehlet A. Johansson, A. Logg, C. Richardson, J. Ring, M. E. Rognes, and G. N. Wells, The FEniCS Project Version 1.5. Archive of Numerical Software, https://fenicsproject.org/ 3 (2015).

Z. I. Botev, J. F. Grotowski, and D. P. Kroese, Kernel density estimation via Diffusion, Ann. Statist 38 (2010), no. 5, 29162957.

P. Chaudhuri and J. S. Marron, Scale space view of of curve estimation, Ann. Statist., 2000, 28 408428. MR1790003.

M. S. Gerber, Predicting crime using Twitter and kernel density estimation, Decision Support Systems 61 (2014), no. 115, 125.

E. L. Glaeser and B. Sacerdote, Why is there more crime in cities?, National Bureau of Economic Research Working Paper 4530, 1996.

F. Gómez, A. Torres, J. Galvis, J. Camargo, and O. Martinez, Hotspot mapping for Perception of Security. Smart Cities Conference (ISC2), IEEE International, 2016.

M. G. Larson and F. Bengzon, The Finite Element Method: Theory, Implementation, and Applications, Springer-Verlag Italia, Milano, 2013.

N. Memon, J. D. Farley, D. L. Hicks, T. Rosenorn, and (Eds.), Mathematical Methods in Counterterrorism, Springer, 2009 Edition.

Y. K. Patarroyo, Mean conservation for density estimation via diffusion using the finite element method, Submitted to Boletín de Matemáticas (2017), https://arxiv.org/pdf/1702.07962.pdf.

S. Salsa, Partial differential equations in action from modelling to theory, Springer-Verlag Italia, Milano., 2008.

B. W. Silverman, Density estimation for statistics and data analysis, Chapman and Hall, London., 1986.

S. Tench, H. Fry, and P. Gill, Spatio-temporal patterns of IED usage by the Provisional Irish Republican Army, European Journal of Applied Mathematics 27 (2016), no. 3, 377-402.

How to Cite

APA

Patarroyo, K. Y., Galvis, J. and Gómez, F. (2018). Preserving data moments in density estimation via diffusion using the finite element method. Boletín de Matemáticas, 25(2), 101–121. https://revistas.unal.edu.co/index.php/bolma/article/view/85491

ACM

[1]
Patarroyo, K.Y., Galvis, J. and Gómez, F. 2018. Preserving data moments in density estimation via diffusion using the finite element method. Boletín de Matemáticas. 25, 2 (Jul. 2018), 101–121.

ACS

(1)
Patarroyo, K. Y.; Galvis, J.; Gómez, F. Preserving data moments in density estimation via diffusion using the finite element method. Bol. Matemáticas 2018, 25, 101-121.

ABNT

PATARROYO, K. Y.; GALVIS, J.; GÓMEZ, F. Preserving data moments in density estimation via diffusion using the finite element method. Boletín de Matemáticas, [S. l.], v. 25, n. 2, p. 101–121, 2018. Disponível em: https://revistas.unal.edu.co/index.php/bolma/article/view/85491. Acesso em: 21 nov. 2024.

Chicago

Patarroyo, Keith Y., Juan Galvis, and Francisco Gómez. 2018. “Preserving data moments in density estimation via diffusion using the finite element method”. Boletín De Matemáticas 25 (2):101-21. https://revistas.unal.edu.co/index.php/bolma/article/view/85491.

Harvard

Patarroyo, K. Y., Galvis, J. and Gómez, F. (2018) “Preserving data moments in density estimation via diffusion using the finite element method”, Boletín de Matemáticas, 25(2), pp. 101–121. Available at: https://revistas.unal.edu.co/index.php/bolma/article/view/85491 (Accessed: 21 November 2024).

IEEE

[1]
K. Y. Patarroyo, J. Galvis, and F. Gómez, “Preserving data moments in density estimation via diffusion using the finite element method”, Bol. Matemáticas, vol. 25, no. 2, pp. 101–121, Jul. 2018.

MLA

Patarroyo, K. Y., J. Galvis, and F. Gómez. “Preserving data moments in density estimation via diffusion using the finite element method”. Boletín de Matemáticas, vol. 25, no. 2, July 2018, pp. 101-2, https://revistas.unal.edu.co/index.php/bolma/article/view/85491.

Turabian

Patarroyo, Keith Y., Juan Galvis, and Francisco Gómez. “Preserving data moments in density estimation via diffusion using the finite element method”. Boletín de Matemáticas 25, no. 2 (July 1, 2018): 101–121. Accessed November 21, 2024. https://revistas.unal.edu.co/index.php/bolma/article/view/85491.

Vancouver

1.
Patarroyo KY, Galvis J, Gómez F. Preserving data moments in density estimation via diffusion using the finite element method. Bol. Matemáticas [Internet]. 2018 Jul. 1 [cited 2024 Nov. 21];25(2):101-2. Available from: https://revistas.unal.edu.co/index.php/bolma/article/view/85491

Download Citation

Article abstract page views

113

Downloads

Download data is not yet available.