Improvement Of Landslide Prediction System Based On Hybrid Neural Networks (Penang Island, Malaysia)
Landslides are one of the most aggressive natural disasters that cause loss of lives and of billions dollars in damages annually worldwide. They pose a threat to the safety of human lives, the environment, resources and property. It is one of the natural disasters that occur quite often in Malays...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://eprints.usm.my/46579/1/Pages%20from%20Improvement%20Of%20Landslide%20Prediction%20System%20Based%20On%20Hybrid%20Neural%20Networks%20%28Penang%20Island%2C%20Malaysia%29%2024%20PAGE.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Landslides are one of the most aggressive natural disasters that cause loss of
lives and of billions dollars in damages annually worldwide. They pose a threat to the
safety of human lives, the environment, resources and property. It is one of the
natural disasters that occur quite often in Malaysia and particularly in Penang Island
during heavy rainy seasons. Numerous researches on landslides studies have been
done based on Penang Island. However, many issues seriously related to landslides
have not been solved yet. These issues include the extraction of new factors which
cause landslides, investigation on the optimum factors which cause landslides and
the generation of an accurate landslide hazard map for Penang island. In addition to
that, the landslide hazard prediction intelligent system, either for Penang Island or for
the entire world is still being investigated up to this date. For that reason, an
intelligent landslide hazard mapping system is proposed. It consists of three stages:
factor extraction, factor selection and Artificial Neural Network (ANNs) as an
analysis tool. Twenty one factors are used in this study where nine factors were
collected from different governmental agents. The rest of the factors (twelve) were
extracted from the Digital Elevation Models (DEM), seven of these factors were
extracted and used for the first time on Penang Island. In the factor selection phase.
six factor selection techniques are employed to select the most important factors in
the landslide prediction. |
|---|