Change search
ReferencesLink to record
Permanent link

Direct link
Lexical Acquisition Made by Machine – A simulation of how a machine learns the meaning of words.
KTH, School of Computer Science and Communication (CSC).
KTH, School of Computer Science and Communication (CSC).
2012 (English)Independent thesis Advanced level (professional degree), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

Learning the meaning of words is a complicated task with many problems. In this study an algorithm to map words to meanings was developed regarding the three problems: handling of sentences (not only singular words), distinguishing the correct of multiple events in a scene and building a lexicon with no entries at the beginning. The aim of this study was to implement an algorithm that would replicate the results of a previous study. The results acquired confirmed the work previously done, the same percentage of word meanings (100%) were learned with equal conditions. To further develop the algorithm problems with words that are spelled identically but mean different things and contexts where events are not describing the utterances said need to be solved. This would make the algorithm more applicable in real world situations.

Abstract [sv]

Att lära sig betydelsen av ord är en väldigt komplicerad uppgift med många problem som behöver lösas. I denna studie utvecklades en algorithm som parar ihop ord med betydelser med avseende på de tre problemen: att kunna hantera meningar (inte bara enstaka ord), att kunna välja ut den rätta händelsen i ett sammanhang samt att kunna lära sig ord utan att tidigare ha kännedom om några ord. Målet med studien var att implementera en algoritm som skulle kunna replikera resultaten i en tidigare rapport på ämnet. De erhållna resultaten fastställde de i ett tidigare arbete, samma andel av betydelser av ord (100%) lärdes in under samma förhållanden. För att ytterligare utveckla algoritmen måste två ytterligare problem lösas: ord som stavas likadant men har olika betydelser och sammanhang där händelserna inte beskriver vad som sades i sammanhanget. Detta skulle göra algoritmen mer användbar i tillämpningar inom ämnet.

Place, publisher, year, edition, pages
Kandidatexjobb CSC, K12003
National Category
Computer Science
URN: urn:nbn:se:kth:diva-131008OAI: diva2:654454
Educational program
Master of Science in Engineering - Computer Science and Technology
Available from: 2013-10-07 Created: 2013-10-07

Open Access in DiVA

No full text

Other links
By organisation
School of Computer Science and Communication (CSC)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 68 hits
ReferencesLink to record
Permanent link

Direct link