Publication Type:

Conference Paper


Proceedings of the International Conference on Data Mining (DMIN), The Steering Committee of The World Congress in Computer Science, Computer Engineering and Applied Computing (WorldComp) (2014)



landslide, pore pressure, rainfall rate, Slope instability, statistical data analysis


Wireless sensor network for landslide detection deployed in Munnar consists of 150 geophysical sensors which are spatially distributed over 20 Deep Earth Probes (DEP) located at different areas in the deployment site. The data received from each of these heterogeneous sensors are mined to retrieve the correlation between the various parameters contributing to landslide, using appropriate statistical methods. This paper presents an architecture which we have developed for automatic data mining of landslide data which will ultimately help in issuing an early warning for occurrence of landslides. Several algorithms were developed towards achieving this objective of effective data analysis of the continuous real time data collected from the deployment field. The results show that the slope instability in a region is dependent not only the intensity of rainfall but also the antecedent rainfall conditions and soil layer parameters. Each of these different algorithms and its results are explained in detail in this paper.

Cite this Research Publication

D. P., B Thottungal, G., Geethalekshmy, C. V., and D Ramesh, M. V., “Automated Statistical Data Mining of a Real World Landslide Detection System”, in Proceedings of the International Conference on Data Mining (DMIN), 2014.