Otkrivanje pogrešaka leksičkog transfera u tekstovima učenika stranog jezika

undergraduate thesis

• Title (english): Detecting Lexical Transfer Errors of Second Language Learners
• Author: Marin Kačan
• Mentor: Jan Šnajder (mentor)
• Committee member: Jan Šnajder (predsjednik povjerenstva)
• Committee member: Domagoj Alagić (član povjerenstva)
• Committee member: Mladen Karan (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2016-07-11, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Automatizirano ispravljanje pogrešaka važan je zadatak obrade prirodnog jezika. Pogreške leksičkog transfera učestala su kod učenika stranih jezika. Najčešći je uzrok transfera višeznačnost riječi. Cilj ovog rada bio je osmisliti model koji otkriva i ispravlja takve pogreške za dva tipa jezičnih relacija: pridjevsko-imeničke i glagolsko-objektne. Implementirana su dva različita modela, od kojih drugi koristi metode nadziranog strojnog učenja, preciznije algoritam regresije pomoću stroja s potpornim vektorima. Jezik implementacije je Python. Ovaj projekt nastao je u suradnji s kineskim sveučilištem Xi'an Jiaotong-Liverpool University, te se oni ustupili skup podataka koji se jednim dijelom koristi u ovome radu.

Abstract (english)

Automated error correction is an important task of natural language processing. Lexical transfer errors are common with L2-learners. The biggest cause of transfer is word polysemy. The goal of this paper was to come up with a model that can detect and correct such errors for two language relation types: adjective-noun and verb-object. Two different models were implemented, the second of which uses supervised learning methods, more precisely the support vector regression algorithm. The model was implemented in Python. This project is carried out in cooperation with Xi'an Jiaotong-Liverpool University, China, who also make the dataset available, a part of which was used in this paper.

• Keywords: obrada prirodnog jezika; automatizirano ispravljanje pogrešaka; leksički transfer; višeznačnost; relacije; stroj s potpornim vektorima; regresija
• Keywords (english): natural language processing; automated error correction; lexical transfer; polysemy; relations; support vector machine; regression
• Language: croatian
• URN:NBN: urn:nbn:hr:168:460306
• 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: 2018-02-06 09:33:42