Information Retrieval with Finnish Case Law Embeddings

Visa fullständig post



Permalänk

http://urn.fi/URN:NBN:fi:hulib-202001211119
Titel: Information Retrieval with Finnish Case Law Embeddings
Författare: Sarsa, Sami
Medarbetare: Helsingfors universitet, Matematisk-naturvetenskapliga fakulteten
Utgivare: Helsingin yliopisto
Datum: 2019
Språk: eng
Permanenta länken (URI): http://urn.fi/URN:NBN:fi:hulib-202001211119
http://hdl.handle.net/10138/310006
Nivå: pro gradu-avhandlingar
Ämne: Tietojenkäsittelytiede
Abstrakt: In this work, five text vectorisation models' capability in embedding Finnish case law texts to vector space for inter-textual similarity computation is studied. The embeddings and their computed similarities are used to create a Finnish case law retrieval system that allows effective querying with full documents. A working web application is presented as a part of the work. The case law data for the work is provided by the Finnish Ministry of Justice, and the studied models are: TF-IDF, LDA, Word2Vec, Doc2Vec and Doc2vecC.


Filer under denna titel

Totalt antal nerladdningar: Laddar...

Filer Storlek Format Granska
grappa-gradu.pdf 1.863Mb PDF Granska/Öppna

Detta dokument registreras i samling:

Visa fullständig post