0% found this document useful (0 votes)
7 views14 pages

Case Study 1

Case study

Uploaded by

Hailey W.
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
0% found this document useful (0 votes)
7 views14 pages

Case Study 1

Case study

Uploaded by

Hailey W.
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/ 14

Case Study 1

Introduction and Course Description

Dr.rer.nat. Anda-Ramona Tănasie

Institut für Informatik, FHWN

February 19, 2024

Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 1 / 11


Presentations

Presenting yourselves
Name, studies, current job
How much experience with Data Science, Python, R
Last Semester lessons leared form Courses connected to Data Science
Why did you choose Data Science?

Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 2 / 11


Presentations

Presenting yourselves
Name, studies, current job
How much experience with Data Science, Python, R
Last Semester lessons leared form Courses connected to Data Science
Why did you choose Data Science?

Last Semester:
Analyse Design
Datenquellen und Datenaufbereitung
Analysewerkzeuge

Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 2 / 11


What is Data Science

What is Data
Information, “something whose value we care about“
Types of Data:
Structured Data: e.g. CSV or "comma-separated values" file
Unstructured Data : not organized, or not categorized, or not in a well-defined
format.
Data Types:
Numbers, Strings ...
Data frames, Series ...

What is Science
Formulating Hypothesis and Validating it
mathematical proof

Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 3 / 11


What is Data Science

What is Data
Information, “something whose value we care about“
Types of Data:
Structured Data: e.g. CSV or "comma-separated values" file
Unstructured Data : not organized, or not categorized, or not in a well-defined
format.
Data Types:
Numbers, Strings ...
Data frames, Series ...

What is Science
Formulating Hypothesis and Validating it
mathematical proof - not possible in most practical problems
statistics, since we deal with uncertainty
Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 3 / 11
Machine Learning

What is Algorithm
Clearly defined series of steps to solve a Task, e.g. Recipe

Algorithms are essential in DS, ML, AI


Learning from Data
computer science, mathematics, statistics, neuroscience, ...

Machine learning
Supervised learning
Unsupervised learning
Semi supervised learning
Reinforcement learning

Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 4 / 11


How it all fits together

AI : creating intelligent machines.


ML : teaching machines to learn without
being explicitly programmed.
DL : use neural networks for learning.
Generative Intelligence: ability of AI systems
to create new, synthetic instances of data

Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 5 / 11


History of AI

AI gained lots of popularity in the last two decades, but it has


actually been around since 1950’s (still one of the newest disciplines)
Turing machine, Turing test, later, in 1956 the name AI was coined
we attempt to understand intelligent entities, and build intelligent
entities as well

What is intelligence?

Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 6 / 11


History of AI

AI gained lots of popularity in the last two decades, but it has


actually been around since 1950’s (still one of the newest disciplines)
Turing machine, Turing test, later, in 1956 the name AI was coined
we attempt to understand intelligent entities, and build intelligent
entities as well

What is intelligence?
understanding intelligence is one of the oldest disciplines
learning, remembering, reasoning...
more difficult as thought, the definitions change constantly

Nowadays:
AI has a huge variety of sub-fields
Narrow vs General AI
Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 6 / 11
Definitions of AI
(Top) Reasoning vs Behaviour (Bottom)
(Left) Human performance vs ideal concept: "rationality" (Right)

Source: Russell and Norvig (2003). Artificial Intelligence: A Modern Approach


(2nd ed.). Prentice Hall, p. 2.
Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 7 / 11
Data Science, Data Analytics and Data Engineering

Data science - quickly evolving


interdisciplinary field

JOURNEY
Goals:
1 Identify your interests
2 Focus on setting one small
reasonable goal at a time
3 Learn coding practices and good
habits for Data Science
4 Statistics, standard algorithms,
paradigms of machine learning
(Classification, Regression,
Clustering, etc...)

Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 8 / 11


Data Science Problem Types
Five main types of questions:
1 How much or how many? (regression)
2 Which category? (classification)
3 Which group? (clustering)
4 Is this weird? (anomaly detection)
5 Which option should be taken? (recommendation)

Parts of a Data Science Project


Data Gathering
Cleaning and preparing the data
Data Exploration, feature selection
Modelling
Visualizations
Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 9 / 11
Course Structure, Content, Objectives
Computational Problem Solving, Statistical Computing
introduce statistical methods and concepts, along with the computational topics
3-5 Projects, combining various topics, technologies and statistical
methods
Case study approach: learning by doing
Real-world data analyses
write code, test it, change it, refine and extend it, generalize it, even sometimes
"start over".
E.g. vectorization, optimizing the code to use the power of R
make educated trade-offs between simplicity, efficiency, generality,
reuse, correct and approximate results, and so on
How to research for a project: Paper presentations
How to present our results: Final project presentations

Workload
Lecture : Home
1 : 2
Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 10 / 11
Generalities
Assumed for this course
install R, RStudio, Python, Jupyter Notebook, VSCode
Tools and libraries for Data Science
Script vs Markdown/Notebooks
Use Documentation
Familiarity with Statistics, Plots ...

Recommandations
R and Python
Stay Organized: Project Folders, Enviroments, Sharing your work,
Proper Communication
Save time: Pipes, Shortcuts, writing short clear code
Build your Portofolio!
Set seed for reproducibility, test your code
Dr.rer.nat. Anda-Ramona Tănasie (FHWN) Case Study 1 February 19, 2024 11 / 11

You might also like