ETD system

Electronic theses and dissertations repository

 

Tesi etd-05302018-110802


Thesis type
Tesi di laurea magistrale
Author
FREDIANI, ILARIA
URN
etd-05302018-110802
Title
POS-tagging improvement for patent analysis: from theory to practice
Struttura
INGEGNERIA DELL'ENERGIA, DEI SISTEMI, DEL TERRITORIO E DELLE COSTRUZIONI
Corso di studi
INGEGNERIA GESTIONALE
Commissione
relatore Prof. Fantoni, Gualtiero
correlatore Dott. Chiarello, Filippo
Parole chiave
  • named entity recognition
  • blockchain
  • chunk extraction
  • improvement
  • patents
  • POS-tagging analysis
  • POS-tagging
  • Natural language processing
Data inizio appello
20/06/2018;
Consultabilità
secretata d'ufficio
Data di rilascio
20/06/2021
Riassunto analitico
The thesis consists of the POS-tagging analysis in order to reduce or eliminate the errors made by the POS-tagger itself. The process is composed by an initial and manual analysis and research of proposals of substitutive tokens or effective text improvements. The proposals are applied to a data set of patents concerning the blockchain: the results are analyzed checking the results of two NLP tasks: chunk extraction and named entity recognition.
File