Awards & Nominations
Project A.G.A.D. has received the following awards and nominations. Way to go!
Global Nominee
Project: Actuation of Geographic Landslide Data(AGAD)
High-Level Project Summary
Every day there are millions of researches and data uploads in search engines, social media, and publishing sites. And if filtered right, it can give valuable input regarding various factors associated with erosion, such as soil moisture, and elevation which will help us identify the landslide vulnerability of different areas. Also reported data from NASA agency can be a source of a machine learning input to predict the possible areas that landslide might occur. After we predict the landslide occurrence a notification /alert can be sent to individuals /residents near the area.
Link to Project "Demo"
Link to Final Project
Detailed Project Description
Our project will be using a Selenium-based web scraper to filter and extract weather-related information on YouTube and online news channels; afterward, the program will automatically convert relevant video links to a CSV file. We will then download the relevant videos, then Python converts their audio into text. Next, we will filter data that contains information related to geographic areas (coordinates), elevation, and weather. Then we will plot them on the Philippine Google map API, together with the prior-mentioned information; this will serve as training data sets, the result of which will be the input for our regression analysis. Lastly, we will plot the data output of the regression analysis on the Google Maps API.
code can be found here: https://bit.ly/3moY1FN or https://bit.ly/3oxUjfA
Kindly refer to the process flowchart for a visual representation of the flow:
Space Agency Data
data used:
Nasa Landslide global catalog points csv
https://power.larc.nasa.gov/data-access-viewer/
Elevation from MERRA2: Average for 0.5 x 0.625 degree lat/lon region = x meters
MERRA2 surface soil wetness
MERRA2 Root zone soil wetness
MERRA2profile soil moisture
MERRA 2 precipitation data corrected
Hackathon Journey
According to Inquirer, the Philippines ranks third in landslide deaths, most of which occur in rural areas, which is why rescue operations cannot guarantee the safety of everyone due to time and distance constraints.
We wanted to be the game changers through data assessment and prediction; our project aims to lessen casualties and improve disaster preparedness for vulnerable areas.
References
Tools used on projects:
python selenium, pandas, sci-kit learn linear regression model, chrome web driver
google maps API
google drive hosting
sources of information
What causes landslides:
https://www.usgs.gov/faqs/what-a-landslide-and-what-causes-one?qt-news_science_products=0#qt-news_science_products
deadly landslides in the PH: https://www.rappler.com/newsbreak/iq/list-deadly-landslides-philippines
Philippines is 3rd most disaster prone country: https://globalnation.inquirer.net/52858/philippines-is-3rd-most-disaster-prone-country-new-study-shows
Tags
#rergession Analysis #websearch #maps #landslide maps #landslideawareness
Global Judging
This project has been submitted for consideration during the Judging process.