Scribd
Upload
39 views46 pages

Lecture6 - Giving Computers Ability To Learn From Data - 2

Machine learning is the process of acquiring new understanding, knowledge, behaviors, skills, values, attitudes, and preferences. The ability of humans and machine learning to learn from data patterns is beneficial because it allows computers and algorithms to gradually improve performance and make data-driven decisions without requiring manual programming. This ability is being used in various fields like computer vision, natural language processing, medicine, and more.

Uploaded by

Radix -
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
39 views46 pages

Lecture6 - Giving Computers Ability To Learn From Data - 2

Machine learning is the process of acquiring new understanding, knowledge, behaviors, skills, values, attitudes, and preferences. The ability of humans and machine learning to learn from data patterns is beneficial because it allows computers and algorithms to gradually improve performance and make data-driven decisions without requiring manual programming. This ability is being used in various fields like computer vision, natural language processing, medicine, and more.

Uploaded by

Radix -
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 46

Fakultas Ilmu Komputer

Learning is the process of acquiring new


understanding, knowledge, behaviors, skills,
values, attitudes, and preferences.

Giving Computers Ability to learn from data


Welcome to the Machine
Learning course
1 - Data vs Knowledge
Why is the ability of humans and machine learning to
learn from data patterns a good thing, and how is this ability
being used in various fields?
Data vs Knowledge

Informasi adalah data Pengetahuan adalah


Data adalah deskripsi yang diolah menjadi informasi yang telah
dari sesuatu dan bentuk yang lebih dikombinasikan dengan
kejadian yang kita berguna dan lebih pemahaman; yang
hadapi. berarti bagi yang lantas melekat di benak
menerimanya. seseorang.
Transform Data Into Knowledge
Building intelligent machines to transform data into knowledge.
 In this age of modern technology, there is one resource that we have in
abundance: (1) a large
amount of structured and
unstructured data.
 In the second half of the twentieth century, machine learning evolved as a subfield
of artificial intelligence that involved the development of (2) self-learning
algorithms to gain knowledge from that data in order to make
predictions.

 Instead of requiring humans to manually derive rules and build models


from analyzing large amounts of data, machine learning offers a more
efficient alternative for capturing the knowledge in data to gradually
improve the performance of predictive models, and make data-driven
decisions.

 Not only is machine learning becoming increasingly important in computer


science research but it also plays an ever greater role in our everyday life.
Instructors
Human Learning vs Pattern Recognition
 Learning is the constant disruption of an old pattern, a breakthrough that
substitutes something old with something new.

 Humans and animals learn with the help of the senses. Manusia dan hewan
belajar dengan bantuan indera.

 Learning helps in identifying and recognizing patterns around us.

The process of pattern recognition involves matching


the information received with the information
already stored in the brain.
 Making the connection between memories and information perceived is a step of
pattern recognition called identification.
Recognizing Patterns
 Pattern recognition requires the
repetition of experience.

 All discoveries and inventions to date


are a result of the pattern
recognition skills of humans.

 Humans have a tendency to see


patterns everywhere.

 They are important when making


comparisons, judgments, and
acquiring knowledge; we tend
to be uneasy with chaos and chance,
just like the foreigner who was
perplexed in the movie theatre.
Why are recognizing and identifying
patterns important in science?

Mengapa mengenali dan mengidentifikasi


pola penting dalam sains?
Face Recognition Technology
AlphaGO Network Architecture

AlphaGo's ability to predict optimal moves in the game is highly


dependent on many factors, including board conditions, previous
experience, and evaluation of current position. The number of
possible moves AlphaGo can generate will vary greatly depending
on the situation in the game.
Chess Network Architecture
Perbandingan antara
kemungkinan gerakan dalam
permainan AlphaGo dan catur
sangat kompleks dan tidak
dapat dinyatakan dalam angka
yang sederhana.

Perbedaan utama antara kedua


permainan tersebut adalah
dalam tingkat kerumitan
dan keragaman gerakan yang
mungkin terjadi.

Dalam catur, terdapat sekitar


10^120 kemungkinan posisi
yang berbeda di seluruh
permainan. Jumlah ini sangat
besar dan hampir tidak dapat
dipahami.
Understanding Machine Learning Algorithms

Manusia belajar dari pengalaman, kita dapat memberikan


kemampuan (ability) ke computer agar dapat belajar dari data.
2 – Types of Learning AI
Different Types of Machine Learning
Types of Learning AI
Types of Learning AI
Machine Learning Algorithms - Classification
Making Predictions
Making predictions about the future with supervised learning.

The main goal in supervised learning is to learn a model from labeled


training data that allows us to make predictions about unseen or future
data. Here, the term supervised refers to a set of samples where the desired
output signals (labels) are already known.
Classification for predicting class labels
 Classification is a subcategory of supervised learning where the goal is to predict
the categorical class labels of new instances based on past observations.
 Those class labels are discrete, unordered values that can be understood as the
group memberships of the instances.
 The previously mentioned example of e-mail-spam detection represents a typical
example of a binary classification task, where the machine learning algorithm
learns a set of rules in order to distinguish between two possible classes: spam and
non-spam e-mail.
Classification vs Regression vs Clustering
The most significant difference
between regression and classification is
that while regression helps predict a
continuous quantity, classification
predicts discrete class labels. There are
also some overlaps between the two
types of machine learning algorithms.
3 – Basic terminology & notations
Column, Feature, Sample & Class

The Iris dataset contains the measurements of 150 iris flowers from three different
species: Setosa, Versicolor, and Virginica. Each flower sample represents one row in
our data set, and the flower measurements in centimeters are stored as columns,
which we also call the features of the dataset
Roadmap for Building AI Systems
A roadmap for building machine learning systems.
Exercise for Students
Exercise #1
Face Recognition dalam computer vision adalah topik yang menarik
dan penting dalam pengembangan sistem yang dapat mengenali
wajah manusia. Students akan diminta untuk menggunakan API
demo untuk mempelajar cara kerja dari face recognition dan
membuat kesimpulan berdasarkan pemahaman mahasiswa.
Input Data Gambar

Perhatikan pada data


gambar di samping,
masing-masing gambar
memilipola dan fitur
wajah yang berbeda. Ada
yang menggunakan kaca
mata, warna kulit yang
berbeda, gambut panjang
dan pendek, usia yang
berbeda, dll.
Face Recognition API
API Demo:
https://www.betafaceapi.com/demo.html
Questions
1) Menurut anda, bagaimana cara sistem ini bisa mengidentifikasi (identify) dan
mengenali (recognize) pola dan fitur wajah?

2) Bangaimana cara untuk mengekstraksi fitur-fitur penting dari wajah, seperti


mata, hidung, mulut, umur, kacamata dll?

3) Menurut anda, fitur/pola wajah apa yang tidak mudah/sulit untuk di extract?

4) Pada khasus seperti apa, pola wajah tidak dapat di identifikasi dan di kenali?

5) Menurut pendapat anda, algoritma apa yang cocok untuk pengenalan wajah
dengan berbagai fitur wajah yang berbeda? Jelaskan kekurangan dan kelebihan
masing-masing algoritma? (machine learning, deep learning, clustering,
classification, regression, etc.)

6) Apakah face recognition sistem bisa bekerja dengan baik pada kondisi malam
hari, dalam hal ini tidak ada cahaya lampu ataupun jenis penerang yang lain?

7) Mengapa gambar ini tidak dapat di kenali pola wajahnya?


Exercise #2
(Text Similarity Algorithm)
Text Similarity Measures
Dalam pemrosesan data berbentuk teks, kita dapat menggunakan fitur text similarity
(ukuran kemiripan teks) untuk mendukung model yang kita bangun. Beberapa
contoh aplikasi text similarity yang dapat digunakan yaitu:

a) Search Engine (Mesin Pencarian)


Dalam mesin pencarian, dapat digunakan text similarity untuk mengukur
kemiripan antara hal yang ingin dicari dengan data yang dimiliki.

b) Chatbot
Ukuran kemiripan digunakan untuk memilih jawaban bot yang tepat
berdasarkan kemiripan pertanyaan atau pernyataan yang ditulis pengguna
dengan database yang ada.

c) Short Answer Scoring


Digunakan untuk menilai suatu jawaban isian singkat dalam suatu soal tes
atau assesment secara otomatis dengan melihat kemiripan antara jawaban
pengerja soal dengan jawaban pada kunci jawaban.

d) Plagiarism Detection
Menggunakan ukuran kemiripan untuk mendeteksi penjiplakan suatu karya
orang lain dan menjadikannya seolah seperti karya sendiri.
K-Nearest Neighbor Classification

Using the Cosine function & K-Nearest Neighbor algorithm, we can determine how
similar or different two sets of items are and use it to determine the classification.
The Cosine function is used to calculate the Similarity or the Distance of the
observations in high dimensional space.
Text Similarity Live Demo

Link: https://similarity-demo.newscatcherapi.com/
Open Google Colab
Import Packages
Create Pandas Dataframe
Cosine Similarity Function
Split Data to Array
Create Dictionary [Unique Words]
Generate Features
Check Similarity
Change Target Text

Jelaskan, mengapa terjadi


error pada saat input
diganti?
END PRESENTATION
Thank you for your attention
Instructor: S – W – T

THANK YOU

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy