Impact of Weighting Factor On Cosine Similarity Based Avalanche Forecasting Model

Full Text (PDF, 387KB), PP.54-60

Views: 0 Downloads: 0

Author(s)

Neha Ajit Kushe 1,* Ganesh Magar 1

1. P. G. Department of Computer Science, S.N.D.T. Women's University, Mumbai - 400049, India

* Corresponding author.

DOI: https://doi.org/10.5815/ijitcs.2019.04.06

Received: 8 Dec. 2018 / Revised: 5 Feb. 2019 / Accepted: 13 Feb. 2019 / Published: 8 Apr. 2019

Index Terms

Cosine Similarity, Forecasting, Nearest Neighbours, Snow Avalanche, Weighting Factor

Abstract

Snow avalanche is an inevitable issue that is faced by mankind residing near the hilly and the mountainous regions. It is a natural disaster that is frequently observed in such terrains. Prediction of these avalanches is crucial for wellbeing of mankind. The concept of using cosine similarity with nearest neighbour is an innovative idea in nearest neighbour based avalanche forecasting model. The results of the model are encouraging, but a need for a mechanism that will provide additional preference to the significant parameters is observed. Present work focuses on the application of weighting factor to the nearest neighbour model with cosine similarity. Use of weighting factor helps in further tuning of the forecasting model. Selection of weighting factors for each parameter is accomplished by considering the effect of each parameter on the avalanche activity. The accuracy of the model is gauged using performance measures - Critical Success Index and Bias and by the changes reflected in the confusion matrix. An increase of 0.1978 and 0.4167 is observed in the values of Critical Success Index after the application of the weights to the forecasting model for dataset combination I and II respectively. The proposed work is implemented using the snow and meteorological data for the Bahang region of Himachal Pradesh, India.

Cite This Paper

Neha Ajit Kushe, Ganesh Magar, "Impact of Weighting Factor On Cosine Similarity Based Avalanche Forecasting Model", International Journal of Information Technology and Computer Science(IJITCS), Vol.11, No.4, pp.54-60, 2019. DOI:10.5815/ijitcs.2019.04.06

Reference

[1]“Avalanches,” 19-Oct-2007. [Online]. Available: himachal.nic.in/WriteReadData/l892s/172_l892s/2-39186387.pdf.

[2]C. Corona and M. Stoffel, “Snow and Ice Avalanches,” in International Encyclopedia of Geography: People, the Earth, Environment and Technology, D. Richardson, N. Castree, M. F. Goodchild, A. Kobayashi, W. Liu, and R. A. Marston, Eds. Oxford, UK: John Wiley & Sons, Ltd, 2017, pp. 1–7.

[3]Schweizer, P. Bartelt, and A. van Herwijnen, “Snow Avalanches,” in Snow and Ice-Related Hazards, Risks and Disasters, Elsevier, 2015, pp. 395–436.

[4]C. Ancey, “Snow avalanches,” in Geomorphological Fluid Mechanics, Springer, 2001, pp. 319–338.

[5]D. M. McClung, “Predictions in avalanche forecasting,” Ann. Glaciol., vol. 31, pp. 377–381, 2000.

[6]D. M. McClung and J. Tweedy, “Numerical avalanche prediction: Kootenay Pass, British Columbia, Canada,” J. Glaciol., vol. 40, no. 135, pp. 350–358, 1994.

[7]N. A. Kushe and G. M. Magar, —An alternative approach with nearest neighbour classifier for forecasting snow avalanche,‖ Int. J. Comput. Eng. Appl., vol. 12, no. 2, pp. 79–85, Feb. 2018.

[8]C. Obled and W. Good, “Recent Developments of Avalanche Forecasting by Discriminant Analysis Techniques: A Methodological Review and Some Applications to the Parsenn Area (Davos, Switzerland),” J. Glaciol., vol. 25, no. 92, pp. 315–346, 1980.

[9]O. Buser, “Avalanche forecast with the method of nearest neighbours: an interactive approach,” Cold Reg. Sci. Technol., vol. 8, no. 2, pp. 155–163, 1983.

[10]O. Buser and W. Good, “Avalanche Forecast: Experience Using Nearest Neighbors,” in Int. Snow Science Workshop, Aspen, Col, Aspen, Colardo, 1984, pp. 109–115.

[11]O. Buser, M. Butler, and W. Good, “Avalanche forecast by the nearest neighbour method,” Int. Assoc. Hydrol. Sci. Publ., vol. 162, pp. 557–569, 1987.

[12]R. Bolognesi, O. Buser, and W. Good, “Local avalanche forecasting in Switzerland: strategy and tools, a new approach,” in International Snow Science Workshop, 1994, pp. 463–472.

[13]K. Kristensen and C. Larsson, “An Avalanche Forecasting Program Based On A Modified Nearest Neighbour Method,” in Proceedings of the 1994 International Snow Science Workshop, Snowbird, Utah, USA, 1994, pp. 22–30.

[14]M. Gassner, H.-J. Etter, K. Birkeland, and T. Leonard, “Nxd2000: An Improved Avalanche Forecasting Program Based On The Nearest Neighbor Method,” in Proceedings Of The International Snow Science Workshop, Big Sky, Montana, USA, 2000, pp. 52–59.

[15]B. Brabec and R. Meister, “A Nearest Neighbor Model For Regional Avalanche Forecasting,” Ann. Glaciol., vol. 32, pp. 130–134, 2001.

[16]L. Mérindol, G. Guyomarc’h, and G. Giraud, “A French Local Tool For Avalanche Hazard Forecasting: Astral, Current State And New Developments,” in Proceedings of the International Snow Science Workshop, Pentiction, Canada, 2002, pp. 105–108.

[17]R. Purves, K. Morrison, G. Moss, and B. Wright, “Cornice- Development Of A Nearest Neighbours Model Applied In Backcountry Avalanche Forecasting In Scotland,” in Stevens, J.R. (Ed.), Proceedings Of International Snow Science Workshop, Penticton, B.C., Canada, 2002, pp. 117–122.

[18]R. . Purves, K. . Morrison, G. Moss, and D. S. . Wright, “Nearest neighbours for avalanche forecasting in Scotland—development, verification and optimisation of a model,” Cold Reg. Sci. Technol., vol. 37, no. 3, pp. 343–355, Nov. 2003.

[19]C. McCollister, K. Birkeland, K. Hansen, R. Aspinall, and R. Comey, “Exploring multi-scale spatial patterns in historical avalanche data, Jackson Hole Mountain Resort, Wyoming,” Cold Reg. Sci. Technol., vol. 37, no. 3, pp. 299–313, Nov. 2003.

[20]C. McCollister, K. Birkeland, K. Hansen, R. Aspinall, and R. Comey, “A probabilistic technique for exploring multi-scale spatial patterns in historical avalanche data by combining GIS and meteorological nearest neighbors with an example from the Jackson Hole Ski Area, Wyoming,” in Proceedings of the 2002 International Snow Science Workshop, Penticton, B.C., Canada, 2002, pp. 109–116.

[21]A. Singh and A. Ganju, “A supplement to nearest-neighbour method for avalanche forecasting,” Cold Reg. Sci. Technol., vol. 39, no. 2–3, pp. 105–113, Oct. 2004.

[22]A. Singh, K. Srinivasan, and A. Ganju, “Avalanche Forecast Using Numerical Weather Prediction In Indian Himalaya,” Cold Reg. Sci. Technol., vol. 43, no. 1–2, pp. 83–92, Nov. 2005.

[23]Amreek Singh and Ashwagosha Ganju, “Artificial Neural Networks for Snow Avalanche Forecasting in Indian Himalaya,” in Proceedings of 12th International Conference of International Association for Computer Methods and Advances in Geomechanics, IACMAG, Goa, India, 2008, vol. 16.

[24]P. Cordy, D. M. McClung, C. J. Hawkins, J. Tweedy, and T. Weick, “Computer assisted avalanche prediction using electronic weather sensor data,” Cold Reg. Sci. Technol., vol. 59, no. 2–3, pp. 227–233, Nov. 2009.

[25]B. Chandra, A. Singh, D. Singh, and A. Ganju, “Features ranking for avalanche forecasting: method and results for north-western Himalaya,” in Proceedings Of The International Snow Science Workshop, 2010.

[26]A. Singh, B. Damir, K. Deep, and A. Ganju, “Calibration of nearest neighbors model for avalanche forecasting,” Cold Reg. Sci. Technol., vol. 109, pp. 33–42, Jan. 2015.

[27]A. Singh, K. Deep, and P. Grover, “A novel approach to accelerate calibration process of a k -nearest neighbours classifier using GPU,” J. Parallel Distrib. Comput., vol. 104, pp. 114–129, Jun. 2017.

[28]A. M. Qamar, E. Gaussier, J.-P. Chevallet, and J. H. Lim, “Similarity Learning for Nearest Neighbor Classification,” 2008, pp. 983–988.

[29]S. S. Sharma and A. Ganju, “Complexities of avalanche forecasting in Western Himalaya—an overview,” Cold Reg. Sci. Technol., vol. 31, no. 2, pp. 95–102, 2000.

[30]H. S. Gusain, V. D. Mishra, and M. R. Bhutiyani, “Winter temperature and snowfall trends in the cryospheric region of north-west Himalaya,” Mausam India, vol. 65, no. 3, pp. 425–432, 2014.

[31]R. Stull, “Wet-Bulb Temperature from Relative Humidity and Air Temperature,” J. Appl. Meteorol. Climatol., vol. 50, no. 11, pp. 2267–2269, Nov. 2011.

[32]A. Battistini, A. Rosi, S. Segoni, D. Lagomarsino, F. Catani, and N. Casagli, “Validation of landslide hazard models using a semantic engine on online news,” Appl. Geogr., vol. 82, pp. 59–65, May 2017.

[33]D. S. Wilks, Statistical methods in the atmospheric sciences, 2 ed. Elsevier.