Otkrivanje informacija iz životopisa zadanih u slobodnom obliku

undergraduate thesis

• Title (english): Information Extraction out of CVs Given In Free Form Text
• Author: Nicolas Assouline
• Mentor: Marin Šilić (mentor)
• Committee member: Marin Šilić (predsjednik povjerenstva)
• Committee member: Petar Afrić (član povjerenstva)
• Committee member: Lucija Šikić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2019-07-12, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Otkad znaju za sebe, ljudi teže prema automatizaciji. Dosad smo uspjeli automatizirati velik broj repetitivnih poslova, i time oslobodili više našeg vremena za produktivnije stvari. Neke poslove je zbog njihove prirode teže automatizirati, a parsiranje i kategorizacija životopisa je jedna od njih. Cilj ovog rada je istražiti područje analize prirodnog jezika i napraviti aplikaciju koja može rastaviti životopis po kategorijama u formatu koji je pogodan za daljnju obradu pomoću računala. Provedena je analiza rezultata na modelima s različitim početnim konfiguracijama i među njima je odabran onaj s najboljim performansama. Model je izgrađen koristeći strojno učenje za označavanje pripadnosti teksta određenoj kategoriji. Ostvareni su zadovoljavajući rezultati i aplikacija (u prosjeku) s otprilike 60\% točnosti grupira tekst u kategorije.

Abstract (english)

Ever since the dawn of man, humanity has been striving towards automation. So far, we have succeeded in automating a large number of repetitive tasks, and in doing so, have freed more of our time for use in more productive endeavors. Some tasks are, due to their nature, harder to automate, and resume parsing is one of them. The aim of this thesis was to explore the field of natural language processing and to create an application that can break down a resume into categories that can be further processed by computers. An analysis of models trained with varying initial configurations has been conducted, and the best among them was chosen. The model used was built and trained using machine learning for sorting text according to the category it belongs to. The results have been satisfactory and the application correctly categorizes text (on average) with close to 60\% accuracy.

• Keywords: Strojno učenje; Obrada prirodnog jezika; NLP; Neuralne mreže; Parsiranje životopisa
• Keywords (english): Machine learning; Natural language processing; Neural networks; Resume parsing
• Language: croatian
• URN:NBN: urn:nbn:hr:168:034883
• 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: 2020-01-07 16:10:06