Skip to main content

AI and Machine Learning (5 cr)

Code: 5000BO52-3002

General information


Enrollment

01.12.2019 - 12.01.2020

Timing

07.01.2020 - 30.04.2020

Number of ECTS credits allocated

5 op

Mode of delivery

Contact teaching

Unit

Engineering and Business

Campus

Kupittaa Campus

Teaching languages

  • English

Seats

15 - 35

Degree programmes

  • Degree Programme in Information and Communication Technology
  • Degree Programme in Information and Communications Technology

Teachers

  • Tapani Ojanperä

Groups

  • PINFOS18
  • VAVA1920
  • PTIVIS18

Objective

After completing the course, the student
* knows the basic terminology related to Artificial Intelligence (AI).

* understands the limits of AI, where and when AI can help solve problems.

* can evaluate what kind of problems are applicable to AI and can use programming libraries to solve easy ones.

Content

· Programming recap
· What is AI?
· AI problem solving
· Statistics for Real World AI
· Machine Learning
· Neural Networks
· Applications

Materials

- Lectures.
- Orange material, https://orange.biolab.si/
- Russell - Norvig: Artificial Intelligence: A Modern Approach (PDF) 3rd Edition
https://readyforai.com/download/artificial-intelligence-a-modern-approach-3rd-edition-pdf/

Teaching methods

Contact hours contain learning material and doing exercises in computer. In the end of the course, student groups (2-3 students) make their own dataset and analyze it. The topic is freely choosed. The goal is to get some interesting and hopefully new knowledge of some phenomenons..

Exam schedules

No exam.

International connections

Students learn to know the terms and .the problems they are applied to with the aid of Orange software. With exercises theoretical issues are trained in. English terms and definitions are the essential part of the studies. The students also utilize videos, tutorials and new learning environments (Orange, MatLab, ACL, Anaconda)

Completion alternatives

None.

Student workload

Contact hours 42 h
Reading lecture material and making exercises. Group task. 93 h
Total 135 h

Content scheduling

- Data mining, classical AI
- Neural networks
- Genetic algorithms
weeks 36 - 50.
We use mainly the Orange tool
.
September: Orange, metrics, linear regression, logistic regression, decision trees.
October: Orange, perceptrons, back propagating algorithms
November: Orange, deep neural networks, self organizing maps, genetic algorithms
December: Real world research.

Evaluation scale

H-5

Assessment methods and criteria

Homework (max 50 p., linear). Continuous assessment [formative assessment, guiding feedback].
Small reporst of the exercises returned to Optima [summative assessment, teacher evaluation]
Learning the concepts of the methods of ML and AI is tested in the final report based on the real world data analysis. (max 20 p.) [summative assessment, teacher evaluation]

Assessment criteria, fail (0)

Student
• does not know basic concepts of machine learning and artificial intelligence.
• returns no solution of the real world data analysis.

Assessment criteria, satisfactory (1-2)

Student
• knows basic concepts of machine learning and artificial intelligence
• understands the basics how to apply a data mining tool to very simple data
• returns a basic solution of the real world data analysis

Assessment criteria, good (3-4)

Student
• knows basic concepts of machine learning and artificial intelligence
• understands a lot of methods how to apply a data mining tool to data
• returns a proper solution of the real world data analysis

Assessment criteria, excellent (5)

Student
• knows many concepts of machine learning and artificial intelligence
• understands a lot of methods how to apply a data mining tool to data
• returns a good solution of the real world data analysis

Qualifications

Basic Programming skills