Tweet Author Gender Identification. PAN 2016 Task

NICVIST, Serghei; BOGATIREVA, Daria; BOBICEV, Victoria

Home
→
Facultatea Electronică şi Telecomunicaţii
→
Conferinţa Internaţională "Telecomunicaţii, Electronică şi Informatică" - ICTEI
→
2018
→
View Item

dc.contributor.author	NICVIST, Serghei
dc.contributor.author	BOGATIREVA, Daria
dc.contributor.author	BOBICEV, Victoria
dc.date.accessioned	2019-06-28T12:45:52Z
dc.date.available	2019-06-28T12:45:52Z
dc.date.issued	2018
dc.identifier.citation	NICVIST, Serghei, BOGATIREVA, Daria, BOBICEV, Victoria. Tweet Author Gender Identification. PAN 2016 Task. In: Telecomunicaţii, Electronică şi Informatică: proc. of the 6th intern. conf., May 24-27, 2018. Chişinău, 2018, pp. 344-347. ISBN 978-9975-45-540-4.	en_US
dc.identifier.isbn	978-9975-45-540-4
dc.identifier.uri	http://repository.utm.md/handle/5014/2810
dc.description.abstract	The paper presents an experiment of tweet’s author gender detection. We used PAN 2016 data and task description and have build an application that decides whether an analysed tweet has been written by man or woman. Multiple texts’ characteristics are used as features in the application, such as: references to pictures, to web pages, to other people, emojis, hashtags and a number of words that are associated with tweets written by women and men respectively. For 100 random tweets we obtained average accuracy 0.61. This is good result although it is not as good as the best one in PAN 2016 task.	en_US
dc.language.iso	en	en_US
dc.publisher	Tehnica UTM	en_US
dc.rights	Attribution-NonCommercial-NoDerivs 3.0 United States	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/3.0/us/	*
dc.subject	user generated content	en_US
dc.subject	text automate analysis	en_US
dc.subject	user profiling	en_US
dc.subject	conținut generat de utilizatori	en_US
dc.subject	analiza automată a textului	en_US
dc.subject	profilarea utilizatorilor	en_US
dc.title	Tweet Author Gender Identification. PAN 2016 Task	en_US
dc.type	Article	en_US