Pristranost u sustavima temeljenim na umjetnoj inteligenciji

undergraduate thesis

• Title (english): Bias in artificial intelligence-based systems
• Author: Ena Maria Maruševec
• Mentor: Mihaela Vranić (mentor)
• Committee member: Mihaela Vranić (predsjednik povjerenstva)
• Committee member: Damir Pintar (član povjerenstva)
• Committee member: Luka Humski (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-08-31, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovaj rad bavi se analizom pristranosti u sustavima temeljenim na umjetnoj inteligenciji. Iako se ti sustavi često smatraju objektivnim, racionalnima i hladno logičnima, neka istraživanja su pokazala da ti sustavi često odražavaju stavove svojih kreatora ili su pak odraz skupova podataka nad kojima su izgrađeni. U radu je dan pregled istraživanja postojećih istraživanja o navedenoj temi. Implementiran je sustav za predlaganje filmova temeljen na algoritmu K-najbližih susjeda. Opisan je način rada navedenog sustava te teorijska podloga. Na primjeru tog sustava pokazani su neki od mogućih načina identifikacije pristranosti te je predloženo rješenje koje bi smanjilo tu pristranost.

Abstract (english)

This paper is regarding the bias in artificial intelligence-based systems. Although these systems are often considered objective, rational and coldly logical, some research has shown that these systems often reflect the attitude of their creators or are a reflection of the data sets on which they are created. The paper provides an overview of previous research on the aforementioned topic. A movie recommendation system based on the K-nearest neighbor algorithm was implemented. The theoretical basis behind the algorithm and how it works are also described. Using the implemented system, some of the possible ways of determining bias are shown, and a solution that would reduce this bias is proposed.

• Keywords: umjetna inteligencija; strojno učenje; pristranost; K-najbližih susjeda; Python
• Keywords (english): artificial intelligence; machine learning; bias; K-nearest neighbours; Python
• Language: croatian
• URN:NBN: urn:nbn:hr:168:141146
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-10-11 09:44:25