The document provides an overview of artificial intelligence (AI) and its subset, machine learning (ML), highlighting various types of learning including supervised, unsupervised, and reinforcement learning. It details the algorithms used in each category, offers applications of ML in forestry, and discusses issues in machine learning such as biases and data quality. The author, Gokul K.S., is a PhD scholar at the Czech University of Life Sciences, with guidance from a supervisor at University College London.

1. Gokul K S, M.Sc., M.Phil.
P.hD. Scholar, Department of Forest Management and Remote Sensing, Faculty
Faculty of Forestry and Wood Sciences, Czech University of Life Sciences (CZU),
(CZU), Prague, Czechia
Supervisor:
Ing. Martin Mokros, Ph.D.
Department of Geography, University College London (UCL), London, UK
1
Introduction to Machine Learning

2. 2
Copyrights @GifDB

3. Artificial Intelligence (AI)
3

4. Artificial Intelligence
■ AI is a branch of Computer Science, that aims to develop computers and
Machines that can act or mimic human like behavours (or Think intelligently)
- Can be Doing some intelligent tasks like Image classification (e.g. person,
objects etc.)
or
Performing complex calculations or analysing the data and providing insights,
Predictions and so on.
4

5. The AI Umbrella
5
Copyrights @Gokul

6. Machine Learning (ML)
Learning : Is the act of acquiring new or reinforcing existing Knowledge,
Behaviours, Skills or Values
Learners can be Humans, birds, animals
Types of Learning :
 Learning Under expert guidance
 Learning guided by Knowledge gained from experts
 Learning by Self
6

7. 7
Copyrights @Adobe Firefly

8. Machine Learning (ML) Cont.
Machine Learning is a sub-set of Artificial Intelligence that enables a system to
learn from data rather than through explicit programing
Or
Machine learning is the Scientific study of algorithms and statistical models that
computer systems use to perform a specific task without using explicit
instructions
8
int a=10, b=20, c=0
c=a+b
print(c)

9. Supervised learning
Learning can be Supervised, Semi-supervised, Unsupervised and reinforcement
 Supervised Learning:
Supervised learning is a learning in which we teach or train the machine using
data which is well labelled that means some data is already tagged with the
correct answer
(Training & Testing)
9

10. 10
Copyrights @Adobe Express Beta

11. Supervised learning Cont.
11
Copyrights @Google

12. Supervised learning Cont.
Supervised learning classified into two categories of algorithms:
 Classification:
A classification problem is when the output variable is a category. Such as “Red”
or “Green” and “Disease” or “No disease”
 Regression:
A regression problem is when the output variable is a real value, such as “Dollars”
or “Weight” / Time series data
12

13. Supervised learning Cont.
Supervised learning algorithms:
 Classification:
o Decision Trees
o Random Forest
o Support Vector Machines (SVM)
o Logistic Regression
o K-Nearest Neighbors (k-NN)
o Naïve Bayes
o Gradient Boosting Machines (GBM)
13
 Regression:
o Linear Regression
o Polynomial Regression
o Lasso Regression
o Decision Trees (for regression)
o Random Forest (for regression)

14. Unsupervised learning
Unsupervised learning is the training of machine using information that is neither
classified nor labelled and allowing the algorithm to act on that information
without guidance.
The task of machine is to group unsorted information according to the
similarities, pattern and differences without any prior training of data
14

15. Unsupervised learning Cont.
15
Copyrights @Google

16. Unsupervised learning Cont.
16
Unsupervised learning classified into two categories of algorithms:
 Clustering:
A clustering problem is where you want to discover the inherent groupings in the
data, such as grouping customers by purchasing behaviour
 Association:
An association rule learning problem is where you want to discover rules that
describe large portions of your data, such as people that buy X also tend buy Y

17. Unsupervised learning Cont.
Unsupervised learning algorithms:
 Clustering:
o K-Means Clustering
o Hierarchical Clustering (Agglomerative
and Divisive)
o DBSCAN (Density-Based Spatial
Clustering of Applications with Noise)
17
 Association:
o Apriori Algorithm
o FP-Growth (Frequent Pattern Growth)
o PrefixSpan Algorithm

18. 18

19. Reinforcement learning
A reinforcement learning algorithm, or agent, learns by interacting with its
environment. The agent receives rewards by performing correctly and penalties
for incorrectly
IVAN PAVLOV’S Classical Conditioning
19
Unconditioned
Stimulus
Unconditioned Response
Neutral Stimulus No Response

20. Reinforcement learning Cont.
20
Unconditioned
Stimulus
Unconditioned Response
Neutral Stimulus
Conditioned Response
During Conditioning
After Conditioning
Conditioned Stimulus

21. Reinforcement learning Cont.
21
Copyrights @Google

22. Reinforcement learning Cont.
Reinforcement learning algorithms:
o Q-Learning
o Deep Q-Networks (DQN)
o Policy Gradient Methods
22

23. 23
Copyrights @Google

24. Application’s in Forestry
o Forest Monitoring and Management: (analyse satellite imagery, drones, or
ground based sensors to monitor forest health and so on)
o Precision Forestry: (Optimal Planting locations, Timber harvesting schedules,
forest inventory management etc.)
o Wildlife Monitoring and Conservations
o Forest Disease and Pest Detection
o Forest Fire Prediction and Management
24

25. Issues in Machine Learning
o Focusing more on Algorithms and theories
o Changing Tools and Packages
o Getting bad predictions and Biases
o Making the wrong assumptions
o Having bad data convert into bad results
25

26. Questions ????
26

27. Thank You 
27
ksg@fld.czu.cz
LinkedIn GitHub