Primjena strojnog učenja u procjeni sklonosti konzumaciji droge

master's thesis

• Title (english): Machine learning approach to estimation of drug consumption risk
• Author: Martina Jurič
• Mentor: Mirjana Domazet-Lošo (mentor)
• Committee member: Mirjana Domazet-Lošo (predsjednik povjerenstva)
• Committee member: Ivica Botički (član povjerenstva)
• Committee member: Marko Horvat (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-07, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Opojna sredstva mogu izazvati razne štetne utjecaje ovisno o količini i učestalosti konzumacije. U sklopu rada, pokušala se odrediti sklonost konzumaciji droge ispitanika pomoću strojnog učenja, na osnovu njihovih demografskih podataka i podataka o osobnosti. Nakon uvoda, dan je kratki opis promatranih opojnih sredstava, opis i vizualizacija korištenog skupa podataka te potom opis modela, postupaka i mjera iz područja strojnog učenja koji su upotrijebljeni u sklopu rješavanja problema. Radi boljeg razumijevanja rezultata klasifikacije, napravljeni su i grafovi ovisnosti nekih demografskih značajki i odgovora ispitanika u originalnom skupu podataka. U nastavku su dani rezultati i kratak osvrt na njih, najprije binarne klasifikacije, a potom i višeklasne klasifikacije za svaku od promatranih droga. Na kraju rada dan je i generalni zaključak te popis korištene literature.

Abstract (english)

Narcotics can cause various harmful effects depending on the amount and frequency of consumption. As part of the work, an attempt was made to determine the propensity to use drugs of the respondents using machine learning, based on their demographic data and personality data. After the introduction, a brief description of the observed narcotics and a description and visualization of the used data set were given followed by a description of the models, procedures and measures in the field of machine learning, which were used for solving the problem. In order to better understand the results of the classification, graphs of the dependence of some demographic characteristics and responses of the respondents in the original data set are made. After that, the results and a brief overview of them are shown, first binary classification and then multiclass classification for each of the observed drugs. At the end, a general conclusion and a list of used literature are given.

• Keywords: droga; strojno učenje; klasifikacija; stroj potpornih vektora; model slučajnih šuma
• Keywords (english): drug; machine learning; classification; support vector machine; random forest
• Language: croatian
• URN:NBN: urn:nbn:hr:168:350024
• 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: 2023-01-13 12:34:55