Scribd
Upload
38 views

Advanced Spatial Analysis Methods - Assignment Questions - Asian Institute of Technology

This is the PDF of my answers

Uploaded by

AdityaGanapathiValiveti
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
38 views

Advanced Spatial Analysis Methods - Assignment Questions - Asian Institute of Technology

This is the PDF of my answers

Uploaded by

AdityaGanapathiValiveti
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

AT 76.

19 Advance Spatial Analysis Methods

Assignment-5 (Submit by April 27)

Q1. Dengue epidemic is almost becoming routine in tropical countries. This is due to one mosquito
Ades Agypti. It is necessary to find which factor has maximum influence on occurrence of dengue or
another terms provides suitable habitat for the Adese Agypti. Please use following data and find out
the Information Values and plot the graph. Which factor has the maximum information value? Which
factor does not relate to habitat of mosquitoes?
Answer:
IV= log (class density/map density)
Class density= ndhf class/nclass
Map desnity= sum (ndhf class)/ sum (nclass)
Map density= 0.3941

All areas are N class Ndhf class Class Class IV


density density/map
density
Class Area Dengue
Affected
Forest 5326.30 23.45 0.0044 0.0116 -1.9355
Water 940.66 82.45 0.0876 0.2227 -0.6522
Builtup 540.44 477.80 0.8837 2.2423 0.3506
Agriculture 8900.32 5607.76 0.6300 1.5985 0.2037
Sum 15707.72 6191.46

IV
0.5

0 class
Forest Water Builtup Agriculture

-0.5

-1

-1.5

-2

-2.5

maximum information value: Built up class


does not relate to habitat of mosquitoes: Forest and Water
Q2. What is Markov Analysis in the context of geospatial sense? Discuss its process and explain how
an application of change assessment can be developed using Markov Model. (200 words)

Markov analysis is a stochastic process that changes over time (temporal changes) in an uncertain
manner in a geographical area (spatial changes). The process of makov, a system that can be in one of
several (numbered) states, and can pass from one state to another each time step according to fixed
probabilities. If a Markov system is in state i, there is a fixed probability, pij , of it going into state j the
next time step, and pij is called transition probability. Stationary Assumption: Transition probabilities
are independent of time ( t).

An application of change assessment using makov model;

- Using Markov chains to analyze changes in wetland trends in arid Yinchuan Plain, China (Zhang et al.,
2010). To predict wetland trend firstly the study markov model built in terms of space-time and its
variables and data. Firstly, land use land cover maps of wetland (1998 and 2006) were created. A
transition probability matrix was built using two wetland distribution maps and GIS. The temporal
dimension and trend is which area (spatial dimension) were predicted using a Markov model. The
prediction model was then tested for relative accuracy and the feasibility of a x2 test.
Q3. Discuss how GIS can be used in developing a Mudslide Risk model. (300 words)
Briefly the use of GIS for MMI; (Mudslide Risk model= Mudslide Index model (MIM)) is based on
combination of Bayesian algorithm and AHP techniques can be activated with spatial data in order to
identify mudslide risk area zonation. Since many GIS supported spatial dynamic data such as rainfall,
vehicle movement, water flow and sedimentation also can analysis to get more accurate results to use as
a more realistic simulation model.

When elaborate the GIS support in developing a MMI model;


Step by steps;
1. To create database using spatial and non-spatial data such field survey, soil, geology, national
economic data, local economic data, text files, field survey GPA and attributes, camera photos,
topo maps, and satellites imageries.
2. To calculate nearest neighbor index (Rn)
3. To find sampling area and sampling points before go to field works using GIS
4. To analysis mudslide distribution based on interpolation techniques such as IDW, Spline and
Kriging in GIS
5. To generate mudslide risk map using satellite RS and surface integrated with Bayesian algorithm.
6. To find appropriate statistical surface for mudslide study which relates slope and aspect.
7. To develop a Mudslide risk model based on information value (IV) approach in GIS
8. To integrate AHP with IV for mudslide risk zonation
9. To simulate the model of mudslide using variables in GIS.
Q4. What is the difference between machine learning and deep learning? Explain with flowcharts
and list some applications.

Explanation: Machine learning is defined “Algorithms that parse data, learn from that data, and then
apply what they’ve learned to make informed decisions” but, deep learning is a sub field of machine
learning. Deep learning structures algorithms in layers to create an “artificial neural network (ANN)” that
can learn and make intelligent decisions on its own.

Key differences:
1. From the way data is presented to the system is different in each method. In machine learning
algorithms almost always require structured data, whereas deep learning networks rely on layers
of the ANN (artificial neural networks).
2. ML use human input to label data but deep learning no use of human, as the nested layers in the
neural networks put data through hierarchies of different concepts
3. Since ML need human inputs it is not good analysis of complex, large scale data but DL is good for
them.
4. DL need more data than ML need.

Application of Deep Learning; Application of Machine Learning;

1. Automatic speech recognition Virtual Personal Assistants (Siri, Alexa)


2. Image recognition / computer vision Traffic prediction
3. Visual art processing Videos Surveillance
4. Natural language processing Social Media Services
5. Drug discovery and toxicology Email Spam and Malware Filtering
6. Customer relationship management in Online Customer Support
direct marketing
7. Bioinformatics Search Engine Result Refining
8. Mobile advertising Product Recommendations
9. Robots technology in military use Online Fraud Detection

You might also like