Alat za interaktivno učenje koncepata umjetne inteligencije i strojnog učenja

master's thesis

• Title (english): Tool for interactive learning of artificial intelligence and machine learning concepts
• Author: Jelena Marković
• Mentor: Ivica Botički (mentor)
• Committee member: Ivica Botički (predsjednik povjerenstva)
• Committee member: Mirjana Domazet-Lošo (član povjerenstva)
• Committee member: Danijel Mlinarić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-02-23, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Rad se fokusira na usporedbu performansi različitih metoda strojnog učenja prilikom zadatka klasifikacije na javno dostupnom skupu podataka za predviđanje srčanih bolesnika. Dan je teorijski opis korištenih metoda kao i odabranog skupa podataka, te analiza rada algoritama nad tim podacima. U sklopu rada izrađena je i interaktivna aplikacija za analizu i vizualizaciju performansi odabranih algoritama. Aplikacija sadrži opis i pregled rezultata za odabranu metodu nad zadanim skupom podataka. Također nudi mogućnosti promjene hiperparametara algoritama, unosa vlastitog skupa podataka te odabir značajki koje će algoritam koristiti. Za izradu aplikacije korišten je programski okvir Dash Plotly koji je baziran na programskom jeziku Python, a pogodan je za izradu interaktivnih aplikacija za analizu podataka.

Abstract (english)

This thesis focuses on the performance comparison of different machine learning methods
on the classification task using a publicly available dataset for predicting heart disease.
It gives a theoretical overview of chosen algorithms, a description of the used dataset, and
an analysis of algorithm performance on it. Also, within the thesis work is developed an
interactive application whose aim is to analyze and visualize the performance of chosen algorithms. The application delivers a description and a visual overview of the results for the
selected method over a given dataset. It also offers possibilities like changing the hyperparameter values of the algorithms, uploading your dataset, and choosing desired features. The application is based on Dash Plotly, a framework based on Python programming language, which is suitable for creating interactive applications for data analysis.

• Keywords: klasifikacija; strojno učenje; analiza; usporedba; logistička regresija; stroj potpornih vektora; algoritam K-najbližih susjeda; naivni Bayesov klasifikator; neuronska mreža; Dash; Ploty; Python; interaktivnost
• Keywords (english): classification; machine learning; analysis; logistic regression; support vector machines; k-nearest neighbors; naive Bayes classificator; convolutional neural network; Dash; Plotly; Python; interactivity
• Language: croatian
• URN:NBN: urn:nbn:hr:168:250164
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-07-13 10:19:44