Please use this identifier to cite or link to this item:
http://hdl.handle.net/2445/97486
Title: | News similarity with natural language processing |
Author: | Parafita Martínez, Álvaro |
Director/Tutor: | Vitrià i Marca, Jordi |
Keywords: | Tractament del llenguatge natural (Informàtica) Intel·ligència artificial Programari Treballs de fi de grau Algorismes computacionals Python (Llenguatge de programació) Natural language processing (Computer science) Artificial intelligence Computer software Bachelor's theses Computer algorithms Python (Computer program language) |
Issue Date: | 28-Jan-2016 |
Abstract: | News articles are pieces of Natural Language that comply with the model of 5W1H, meaning, they should answer to the following six questions: What, Who, Where, When, Why and How. This project takes advantage of that assumption to create an algorithm capable of building a representation of a news article and a distance between such representations for any pair of politics news. With that knowledge, a global dis- tance between entries based on similarity of content is built. That algorithm is assessed in comparison with the topic modeling algorithm Latent Dirichlet Allocation (LDA). Applications of the system with their corresponding visualisations are presented too. |
Note: | Treballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2016, Director: Jordi Vitrià i Marca |
URI: | http://hdl.handle.net/2445/97486 |
Appears in Collections: | Treballs Finals de Grau (TFG) - Enginyeria Informàtica Programari - Treballs de l'alumnat |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
memoria.pdf | Memòria | 2.36 MB | Adobe PDF | View/Open |
codi_font.zip | Codi font | 2.4 MB | zip | View/Open |
This item is licensed under a Creative Commons License