Author profiling from images using 3D convolutional neural networks

Eduardo Valdez-Rodríguez; Hiram Calvo; Edgardo Felipe-Riverón

Author profiling from images using 3D convolutional neural networks

Eduardo Valdez-Rodríguez, Hiram Calvo, Edgardo Felipe-Riverón

Centro de Investigación en Computación (CIC)

Research output: Contribution to journal › Conference article › peer-review

2 Scopus citations

Abstract

With this work we participate in the competition on the author profiling task on the MEX-A3T track at IberEval 2019. Author profiling task aims to identify gender, occupation and location from images or text of Mexican Twitter users. We propose a 3D Convolutional Neural Network for solving this task, using visual information, in this case images extracted from the user’s profile which are grouped to create a unique input of each user.

Original language	English
Pages (from-to)	508-514
Number of pages	7
Journal	CEUR Workshop Proceedings
Volume	2421
State	Published - 2019
Event	2019 Iberian Languages Evaluation Forum, IberLEF 2019 - Bilbao, Spain Duration: 24 Sep 2019 → 24 Sep 2019

Keywords

3D CNN
Author profiling
Visual information

Cite this

@article{be9d61f23eee4ba3b32970ccf316aea7,

title = "Author profiling from images using 3D convolutional neural networks",

abstract = "With this work we participate in the competition on the author profiling task on the MEX-A3T track at IberEval 2019. Author profiling task aims to identify gender, occupation and location from images or text of Mexican Twitter users. We propose a 3D Convolutional Neural Network for solving this task, using visual information, in this case images extracted from the user{\textquoteright}s profile which are grouped to create a unique input of each user.",

keywords = "3D CNN, Author profiling, Visual information",

author = "Eduardo Valdez-Rodr{\'i}guez and Hiram Calvo and Edgardo Felipe-River{\'o}n",

year = "2019",

language = "Ingl{\'e}s",

volume = "2421",

pages = "508--514",

journal = "CEUR Workshop Proceedings",

issn = "1613-0073",

publisher = "CEUR-WS",

}

TY - JOUR

T1 - Author profiling from images using 3D convolutional neural networks

AU - Valdez-Rodríguez, Eduardo

AU - Calvo, Hiram

AU - Felipe-Riverón, Edgardo

PY - 2019

Y1 - 2019

N2 - With this work we participate in the competition on the author profiling task on the MEX-A3T track at IberEval 2019. Author profiling task aims to identify gender, occupation and location from images or text of Mexican Twitter users. We propose a 3D Convolutional Neural Network for solving this task, using visual information, in this case images extracted from the user’s profile which are grouped to create a unique input of each user.

AB - With this work we participate in the competition on the author profiling task on the MEX-A3T track at IberEval 2019. Author profiling task aims to identify gender, occupation and location from images or text of Mexican Twitter users. We propose a 3D Convolutional Neural Network for solving this task, using visual information, in this case images extracted from the user’s profile which are grouped to create a unique input of each user.

KW - 3D CNN

KW - Author profiling

KW - Visual information

UR - http://www.scopus.com/inward/record.url?scp=85071193876&partnerID=8YFLogxK

M3 - Artículo de la conferencia

AN - SCOPUS:85071193876

SN - 1613-0073

VL - 2421

SP - 508

EP - 514

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

T2 - 2019 Iberian Languages Evaluation Forum, IberLEF 2019

Y2 - 24 September 2019 through 24 September 2019

ER -

Author profiling from images using 3D convolutional neural networks

Abstract

Keywords

Other files and links

Fingerprint

Cite this