Professional Documents
Culture Documents
Abstract 1. Introduction
Haphazard development activities on mountain slopes Landslides are the geological incident which includes
and inadequate attention to construction aspects have led widely spread land movement resulting meticulous
to the increase of landslides and consequently sustaining damages to the people and their belongings.
damage to lives and infrastructure. Nearly 3275 sq.km o f Fundamentally the landslide transpired when a part of a
area spread over the Ratnapura District, seems to be natural slope is not capable of bearing its weight. The
highly prone to landslides and mass wastage o f 2178 gravity is the fundamental driving force of the landslide
sq.km. Landslides occur in many regions o f Ratnapura refuse flow relying on the slope of the area. Landslide
district and nearly 90 deaths have been reported happens when the stability of the slope changes from a
according to National Research Building Organization stable to an unstable state. In the last decades, there was a
(NBRO) in 2017. Most landslides or potential failures considerable increase in landslide frequency, in accord
could be predicted fairly accurately i f proper with the climatic changes, improper land uses and the
investigations were performed in time. The primary expansion of urbanized areas in the world. Thus,
objective o f this study is landslide-hazard mapping and landslide detection is a crucial requirement in pre and
risk evaluation to determine the real extent, timing, and post-disaster, hazard analysis processes.
severity o f landslide processes in Ratnapura district. The recognition of landslide susceptibility is
Such knowledge will provide the most significant benefit important to get some preventive and control actions and
to government officials, consulting engineering firms, give some early warnings to reduce or mitigate hazards
and the general public in avoiding the landslide hazard impacts. Most developed countries in the world apply the
or in mitigating the losses. Hybrid Machine Learning latest tools for landslide prediction. But Sri Lanka being
techniques can be used to develop prediction models an underdeveloped country cannot meet the expenses for
using existing data. Ensemble approach based on such technologies. So, it is necessary to bring a research
Support Vector Machine (SVM), Naïve Bayes model were study for landslide predictions to give early warnings
combined and implemented fo r the final prediction. This which lead save the lives of naive people.
study possesses a strong capability to predict landslides Machine learning is a sophisticated fusion of applied
by causative factors, slope, land use, elevation, geology, mathematics and computational intelligence. It focuses
soil materials and triggering factor; rainfall was on ‘training’ an algorithm to probe for and leam from
extracted and applied to the machine learning data structure robust enough to make predictions; even
algorithms. This research introduces a novel architecture without predecessor knowledge of the structure. In
to produce a more relevant and accurate prediction o f Machine Learning perspective, a majority of studies
the landslide vulnerability within the study area. discussed aspect related to Artificial Neural Network
Moreover, it was revealed that all o f the factors had (ANN), Support Vector Machine (SVM), Decision tree,
relatively positive relationship with occurrence o f Classifying and Clustering, Bayesian, WebGIS and
landslides. An improvement in hazard monitoring, General data mining techniques.
accuracy o f early warning and disaster mitigation is Even though these models have been applied
documented. successfully and efficiently in landslide susceptibility
assessment, no model is perfect. Therefore, the
Keywords: Landslide, Support Vector Machine (SVM), improvement in these models is needed to achieve
Naïve Bayes, Hydrological, Rainfall desirable results. The performance of landslide models
can be enhanced by using feature selection and ensemble.
Whereas, the Ensemble frameworks that combine
multiple classifiers to improve the performance of
4. Objectives
The natural hazards are beyond human control, but
their destruction can be reduced if prediction mechanisms
are carried out in advance. Researchers worldwide are
having a great pace to develop early prediction
mechanisms for such natural hazards. It was hard to use
traditional mathematical methods for analyzing. Figure 2. (a) Kiribathgala,
The paper aims in presenting an architecture using in Ratnapura district, Sri
Lanka, Monday, May 29,
Ensemble approach relying on an integration of SVM
2017, (b) Ratnapura,
and Naïve Bayes which possess a strong capability to Kalawana and Ayagama
predict landslides and analyze the climate variability areas (c) Colombo Hatton
(rainfall and temperature). main road due to rock falls
5.2.1. Required data Naive Bayes classiher is one of the simplest soft
computing methods which is based on the Bayesian
To determine the tributary characteristics of the study theory and the maximum posterior hypothesis [12].
area a) Digital topographic maps of the study area Naive Bayes classiher uses a statistical hypothesis that all
(1:10000) and b) Digital elevation model (DEM) - values of numeric attributes are independent and
satellite images of specific areas were obtained. normally distributed in each class. Naive Bayes classiher
has been applied effectively in many helds such as
T ab le l. Datausedandsources. medical diagnosis, and management. However, its
application is still limited to landslide problems [13].
D a t a u se So u rces
5.3.2. SupportVector Machine
Rainfall Metrology Department
Support Vector Machine is primary a classier method
Soil Materials Irrigation Department
that performs the classification tasks by constructing
Geology Survey Department
hyperplanes in a multi-dimensional space that separates
LandUse LUPP Department cases of different class labels. SVM supports both
Landslide & non Landslide NBRO regression method and classification technique tasks and
Soil Texture Survey Department can handle multiple continuous and categorical variables
Distance from road, river NBRO [14].
Influence of constructions NBRO
5.3.3. The novel classifier ensemble model
To analyze the climate variability (rainfall and
In this study, the novel ensemble classiher model was
temperature), monthly rainfall and temperature data in
generated by the combination of Naive Bayes classiher
the past, recent five years (2012-2017)
and SVM ensemble. SVM ensemble was hrst applied to
To study the slope aspect, angle within the landslide create the subsets of training[15]. After that, Naive Bayes
are National Building Research Organization | Hazard
classiher was used to construct base classihers from these
Zonation Map was referred.
subsets for classihcation. Methodological how chart of
the novel classiher ensemble model is shown in Figure 3.
5.2.2. Software tools and models
■ Arc GIS 10.4 version
■ Digital Elevation Model (DEM)
Classifier Ensemble model
Bagging
Figure 4. Reference architecture of prediction model.
Bagging uses to implement similar learners on small
sample populations and then takes a mean of all the
6. Results and discussion predictions.
The natural disasters are beyond human control, but their
destruction can be reduced if prediction mechanisms are
carried out in advance. Researchers worldwide are
having a great place to develop an application of the
landslide prediction model an exploration always, and it
includes many things to be researched further. The SVM
apply, and the results use to produce landslide prediction
model of the study areas.
Landslide Locations in Ratnapura District, Sri Lanka
Boosting
This technique uses to adjust the weight of an
observation based on the last combined classification. If
an observation was classified incorrectly, it tries to
increase the weight of these observations.
Legend
♦ LS_Ratnapura
I Rainapura_LH3l
8. Acknowledgement
The access to Hazard Zonation map of Ratnapura
District from the Sri Lanka National Research Building
Organization (NBRO), Director General, Department of
Metrology, Dr. Mr. R. M. S. Bandara has been
Figure 8. Stacking process, to decrease in either bias acknowledged, Mr. A. L. K. Wijemanna, Director
or variance error depending on the combining (Computer, Research, Climate change, International
learner. Affairs) was supported to collect Rainfall Data and also
Director General, Department of Land Use & Policy
Ensemble approach produces more accurate solutions Planning, Mrs. A. S. Illangamge helped to collect Land
than a single model would. When we are trying to predict Use data and my sincere gratitude to the Sabaragamuwa
the target variable using any machine learning Ensemble University of Sri Lanka for the encouragement.
technique, the leading causes of difference in actual and
predicted values are noise, variance, and bias. Ensemble 9. References
approach has helped to reduce these factors, except noise,
which is an irreducible error. [1] Pham, B. T.,Bui, D.T.,Dholakia, M.B., Prakash, I., Pham,
A landslide prediction model should be able to make H.V., Mehmood, K. and Le, H.Q., “A novel ensemble classifier
an adequate prediction of possible landslide areas to of rotation forest and Naïve Bayer for landslide susceptibility
improve the hazard monitoring, accuracy of early assessment at the Luc Yen district, Yen Bai Province (Viet
Nam) using GIS,” Geomatics, Nat. Hazards Risk, vol. 8, no. 2,
warning and disaster mitigation by collecting data of the
pp. 649-671,2017.
triggering factors of the disaster. Data collected for this
research may provide a suitable database for hazard [2] Pham, B. T. , Shirzadi, A., Tien Bui, D. , Prakash, I. and
forecasting and future studies. Dholakia, M. B., “A hybrid machin/e learning ensemble
approach based on a Radial Basis Function neural network and
Rotation Forest for landslide susceptibility modeling: A case
7. Conclusion study in the Himalayan area, India,” Int. J. Sediment Res., vol.
Landslide susceptibility assessment have been done 33, no. 2, pp. 157-170,2018.
in the Ratnapura District, Sri Lanka using the novel [3] Government of Sri Lanka, World Bank, UN Sri Lanka, and
ensemble classiher model which is a combination of Global Facility for Disaster Reduction and Recovery, Sri Lanka
Naive Bayes classiher and SVM. Naive Bayes is an Rapid Post Disaster Needs Assessment: Floods and Landslides,
effective classiher. Analysis results show the novel May 2017, no. May. 2017.
classiher ensemble model has the best degree of ht to [4] Kadavi, P. R. , Lee, C. W., and S. Lee, “Application of
landslide susceptibility assessment compared to other ensemble-based machine learning models to landslide
models. This study identihed factors that may be susceptibility mapping,” Remote Sens., vol. 10, no. 8, pp. 1-18,
involved in landslides, and the results and methods that 2018.
can be used for landslide predicting in other regions [5] Gariano, S. L. and Guzzetti, F., “Landslides in a changing
beyond the study areas. Landslide prediction models can climate,” Earth-Science Rev., vol. 162, no. August 2016, pp.
help implement a guide for planning the mass evacuation 227-252, 2016.
of residents in the case of a landslide, and also to prevent
[6] Pham, B. T. and Prakash, I., “A novel hybrid model of
or reduce the disruptive impacts of a natural disaster on Bagging-based Naïve Bayes Trees for landslide susceptibility
surrounding communities. assessment,” Bull. Eng. Geol. Environ., pp. 1-15, 2017.
Therefore, the present study proposed an
Architecture for predicting landslide occurrences in the [7] Terlien, M. T. J., “The determination of statistical and
areas susceptible to these phenomena in the Ratnapura deterministic hydrological landslide-triggering thresholds,”
Environ. Geol., vol. 35, no. 2-3, pp. 124-130, 1998.