eprintid: 48255
rev_number: 11
eprint_status: archive
userid: 1482
importid: 105
dir: disk0/00/04/82/55
datestamp: 2023-09-25 08:45:47
lastmod: 2024-11-04 12:18:51
status_changed: 2024-11-04 12:18:51
type: monograph
metadata_visibility: show
creators_name: Gadat, Sébastien
creators_name: Villeneuve, Stéphane
creators_id: sebastien.gadat@tse-fr.eu
creators_id: stephane.villeneuve@tse-fr.eu
creators_idrefppn: 080889433
creators_idrefppn: 061507105
creators_affiliation: Toulouse School of Economics;Institut Universitaire de France
creators_affiliation: Toulouse School of Economics
creators_halaffid: 1002422
creators_halaffid: 1002422
title: Parsimonious Wasserstein Text-mining
ispublished: pub
subjects: subjects_ECO
abstract: This document introduces a parsimonious novel method of processing textual data based on the NMF factorization and on supervised clustering withWasserstein barycenter’s to reduce the dimension of the model. This dual treatment of textual data allows for a representation of a text as a probability distribution on the space of profiles which accounts for both uncertainty and semantic interpretability with the Wasserstein distance. The full textual information of a given period is represented as a random probability measure. This opens the door to a statistical inference method that seeks to predict a financial data using the information generated by the texts of a given period.
date: 2023-09
date_type: published
publisher: TSE Working Paper
official_url: http://tse-fr.eu/pub/128497
faculty: tse
divisions: tse
keywords: Natural Language Processing
keywords: Textual Analysis
keywords: Wasserstein distance
keywords: clustering
language: en
has_fulltext: TRUE
view_date_year: 2023
full_text_status: public
monograph_type: working_paper
series: TSE Working Paper
volume: 23-1471
place_of_pub: Toulouse
pages: 20
institution: Université Toulouse Capitole
department: Toulouse School of Economics
book_title: TSE Working Paper
oai_identifier: oai:tse-fr.eu:128497
harvester_local_overwrite: department
harvester_local_overwrite: pending
harvester_local_overwrite: creators_idrefppn
harvester_local_overwrite: creators_halaffid
harvester_local_overwrite: abstract
harvester_local_overwrite: place_of_pub
harvester_local_overwrite: creators_affiliation
harvester_local_overwrite: institution
harvester_local_overwrite: pages
harvester_local_overwrite: creators_id
oai_lastmod: 2024-04-29T08:30:06Z
oai_set: tse
site: ut1
citation:   Gadat, Sébastien <https://www.idref.fr/080889433> and Villeneuve, Stéphane <https://www.idref.fr/061507105>  (2023) Parsimonious Wasserstein Text-mining.  TSE Working Paper, n. 23-1471, Toulouse     
document_url: https://publications.ut-capitole.fr/id/eprint/48255/1/wp_tse_1471.pdf