Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Speaker Deck
PRO
Sign in
Sign up
for free
Machine Learning 101
Ali Akbar S.
December 18, 2017
Education
1
68
Machine Learning 101
Ali Akbar S.
December 18, 2017
Tweet
Share
More Decks by Ali Akbar S.
See All by Ali Akbar S.
aliakbars
0
12
aliakbars
1
38
aliakbars
2
170
aliakbars
0
23
aliakbars
0
46
aliakbars
0
39
aliakbars
0
30
aliakbars
0
30
aliakbars
1
59
Other Decks in Education
See All in Education
weavertheatre
0
180
vlachoca1
0
130
kaityo256
6
3.1k
atp
0
130
salvemaria
0
160
learnenergy2
0
150
signer
0
330
signer
0
220
yasslab
PRO
4
64k
kuro_wd
0
120
signer
0
240
r0mashka
0
210
Featured
See All Featured
lemiorhan
626
42k
mza
80
4.1k
jmmastey
8
530
wjessup
338
16k
destraynor
146
19k
erikaheidi
13
4.2k
chrislema
231
16k
geeforr
332
29k
marktimemedia
6
330
chrislema
173
14k
sugarenia
233
830k
samanthasiow
56
6.3k
Transcript
Machine Learning 101 Ali Akbar Septiandri Universitas Al Azhar Indonesia
Previously...
Cross Industry Standard Process for Data Mining (CRISP-DM)
Data Science Venn Diagram
What is the role of machine learning algorithms?
“Fundamentally, machine learning involves building mathematical models to help understand
data.” - Jake VanderPlas
Tasks in Machine Learning 1. Predicting stock price 2. Differentiating
cat vs. dog pictures 3. Spam identification 4. Community detection 5. Mimicking famous painting style 6. Mastering the game of go and chess 7. etc.
Task Categories 1. Supervised learning a. Predicting stock price b.
Differentiating cat vs. dog pictures c. Spam identification 2. Unsupervised learning a. Community detection b. Mimicking famous painting style 3. Reinforcement learning a. Mastering the game of go and chess
- Iris Dataset - by R.A. Fisher (1936) - 4
attributes: sepal length, sepal width, petal length, petal width - 3 labels: Iris Setosa, Iris Versicolour, Iris Virginica Let’s take an example dataset...
None
None
None
None
None
Nearest Neighbour - Finding the closest reference - What does
it mean by “closest”? - Humans comprehend visualisations very well - Can computers do the same?
At the lowest level, computers only understand 0 or 1
Euclidean Distance
Euclidean Distance
Are you sure?
1. Find some k closest references 2. Use majority vote
3. We need to compute pairwise distances k-Nearest Neighbours
None
Conventional statistics can not do that
We need high computational power
What if we only want to see the subgroups in
the data?
Clustering - Finding subgroups in the data - Your neighbours
in the same housing complex regardless of their class - Unsupervised learning
None
k-Means Clustering
k-Means Clustering 1. Uses Euclidean distance as well 2. k
= number of clusters 3. Centroids to represent clusters
None
None
None
Deep Learning
None
Digit Recognition MNIST Dataset
Classifying objects from pictures [Krizhevsky, 2009]
None
None
A neural network [Nielsen, 2016]
Logistic Regression y = σ(w 0 + w 1 x
1 )
Predicting traffic jams from CCTV pictures
Mimicking famous paintings
None
Other Machine Learning Algorithms
Naive Bayes
Decision trees
Linear regression with polynomial basis functions
“No free lunch”
Thank you