umu.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Semantic Similarity Evaluation using Convolution Kernels and Word Embeddings
Umeå University, Faculty of Science and Technology, Department of Computing Science.
2018 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Human language allows for the same core meaning to be expressed in several different ways. The evaluation of similarity in meaning between two sentences are of interest for many natural language processing applications. However, an analysis of the syntactic components of the sentences, such as an application of a convolution kernel onto syntactic parse trees, would not be concerned with semantic meaning. The addition of word embedding functionality could benefit the kernel and derive the meaning from the semantic properties of the words involved.This thesis proposes that a convolution kernel used in conjunction with word embeddings would improve the classification performance of semantically similar sentences. Two such combinations are suggested and tested as classifiers against that of a basic convolution kernel on syntactic trees. The results of this experiment indicate that the performance does improve by utilizing the word embeddings. Unfortunately, the accuracy of the developed method is worse than other suggested approaches to the task.

Place, publisher, year, edition, pages
2018. , p. 47
Series
UMNAD ; 1152
National Category
Engineering and Technology
Identifiers
URN: urn:nbn:se:umu:diva-152025OAI: oai:DiVA.org:umu-152025DiVA, id: diva2:1250451
Educational program
Master of Science Programme in Computing Science and Engineering
Supervisors
Examiners
Available from: 2018-09-24 Created: 2018-09-24 Last updated: 2018-09-24Bibliographically approved

Open Access in DiVA

No full text in DiVA

By organisation
Department of Computing Science
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 69 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf