Otkrivanje kompetencija iz životopisa zadanih u slobodnom obliku

master's thesis

• Title (english): Extraction of Hard Skills from Free-form Resumes
• Author: Sebastian Dovičin
• Mentor: Marin Šilić (mentor)
• Committee member: Marin Šilić (predsjednik povjerenstva)
• Committee member: Klemo Vladimir (član povjerenstva)
• Committee member: Goran Delač (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

U životopisu se nalaze čovjekove kompetencije, koje pobliže predstavljaju sposobnosti osobe. Koristeći alate za obradu prirodnog jezika, moguće je izdvojiti samo kompetencije iz svakog životopisa. Prirodni jezik je složen, stoga za programsko ostvarenje treba koristiti napredne tehnologije poput neuronskih mreža. Svaku riječ prije slanja u neuronsku mrežu potrebno je pretvoriti u jedinstveni vektor brojeva zbog prirode rada računala s brojevima. Sustav uči nad označenim skupom fraza i pripadnih konteksta te analizira novi životopis i za svaku frazu iz životopisa procijeni kolika je vjerojatnost pripadnosti nekoj od kategorija. Sustav mora imati početni vokabular i sadrži puno elemenata koji mogu poboljšati točnost.

Abstract (english)

There is a list of person's skills in his resume. Those skills represent person's abilities. It is possible to extract only skills from every resume by using Natural Language Processing. Natural language is complex, so advanced technologies, like neural networks, must be used in extraction system. Before sending to neural network, every word must be converted to unique vector of numbers because computers work best with numbers. The system is trained using marked set of phrases and associated context. When new resume is given to it, the system calculates probability of affiliation to a category for each phrase. The system must learn from an initial vocabulary, and it contains a lot of parameters by which the system can be fine-tuned to improve accuracy.

• Keywords: kompetencija; životopis; strojno učenje; neuronska mreža; obrada prirodnog jezika
• Keywords (english): skill; resume; cv; machine learning; neural network; natural language processing
• Language: croatian
• URN:NBN: urn:nbn:hr:168:289562
• 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 09:56:12