Přednáška "Classifying Medical Data with Neural Nets and Inductive Logic Programming" Bettina Finzel (Researcher at Cognitive Science Group, Univ. of Bamberg)
|Začátek:||úterý 22. října 2019, 16:00|
|Konec:||úterý 22. října 2019, 17:45|
|Místo konání:||SB 202|
|Kontaktní osoba:||Doc. Ing. Tomáš Kliegr, Ph.D.|
Machine learning can be used to make predictions on data for various use cases. In this session we will have a look at medical data. You will apply two different machine learning approaches to decide, whether a person has the flu or not and whether a microscopy scan shows healthy or cancerous tissue. In this practical session, you will learn about the basics of neural nets and inductive logic programming. First, we will implement a simple neural net in R to make predictions on the flu data set. You will learn how to train, test and plot the net. In the second part of the session, we will apply inductive logic programming to make a diagnosis for the microscopy scans. We will discuss requirements that should be fulfilled by machine learning approaches in the medical domain, such as transparency and comprehensibility.
Code snippets are provided to you. Experience with machine learning or R is not required.