Algoritmi strojnog učenja primijenjeni na zadanom skupu podataka

undergraduate thesis

• Title (english): Machine learning algorithms applied to a given data set
• Author: Anja Gelo
• Mentor: Tomislav Stipančić (mentor)
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2023-03, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Mechanical Engineering

Abstract

U svijetu podataka i velike količine informacija algoritmi umjetne inteligencije postaju sve značajniji. U ovisnosti o vrsti i strukturi podataka te metodi strojnog učenja algoritmi se koriste prilikom predviđanja, odlučivanja, klasifikacije, prepoznavanja i slično. U ovom radu opisane su i objašnjene tri klasifikacijske neparametarske metode strojnog učenja koje pripadaju nadgledanom učenju – stroj potpornih vektora, algoritam k-najbližih susjeda te stablo odluke. Sva tri algoritma su primijenjena na zadanom skupu podataka u programskom jeziku Python te open-source web aplikaciji Jupyter Notebook. Prije same primjene algoritama na zadani skup podataka, na podatke je bilo potrebno primijeniti funkcije obrade podataka – funkcije čitanja, funkcije čišćenja, funkcije skaliranja te funkcije vizualizacije. Algoritmi su uspoređeni po validacijskom kriteriju preciznosti.

Abstract (english)

In the world of data and large amounts of information, artificial intelligence algorithms are becoming increasingly important. Depending on the type and structure of the data and the machine learning method, algorithms are used for prediction, decision-making, classification, recognition, and the like. In this paper, three classification non-parametric machine learning methods that belong to supervised learning are described and explained - the support vector machine, the k-nearest neighbors algorithm and the decision tree. All three algorithms were applied to the given data set in the Python programming language and the open-source web application Jupyter Notebook. Before applying the algorithms to the given data set, it was necessary to apply data processing functions to the data - reading functions, cleaning functions, scaling functions and visualization functions. Algorithms were compared according to the validation criterion of precision.

• Keywords: strojno učenje; algoritam; SVM; stroj potpornih vektora; stablo odluke; klasifikator k-najbližih susjeda; umjetna inteligencija; klasificiranje podataka
• Keywords (english): machine learning; algorithm; SVM; machine supported vectors; decision tree algorithm; K-NN; classified data; artificial intelligence
• Language: croatian
• URN:NBN: urn:nbn:hr:235:397812
• Study programme: Title: Mechanical Engineering; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka strojarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka strojarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2023-02-23 10:23:06