Κατεύθυνση Ψηφιακά Μέσα Επικοινωνίας και Περιβάλλοντα Αλληλεπίδρασης
Library of the Faculties of Political Science and Public Administration, Communication and Mass Media Studies, Turkish and Modern Asian Studies, Sociology

2018-11-29

2018

Oikonomopoulos Charalampos

Κωνσταντίνος Μουρλάς, Επίκουρος Καθηγητής, Τμήμα Επικοινωνίας και Μέσων Μαζικής Ενημέρωσης, ΕΚΠΑ

Ανάλυση Συναισθήματος με χρήση τεχνικών εξόρυξης άποψης σε σχόλια στο YouTube και εφαρμογή τους στο Marketing

Greek

YouTube comments Sentiment Analysis using Opinion Mining Techniques & application in Marketing

Social Media users share through their comments their opinions and express their feelings creating big data, of which the analysis constitutes a considerably demanding process. Aforementioned data provides interesting conclusions concerning the consumers’ opinions and for this reason their extraction and evaluation is necessary for various purposes in the sales development.
The scope of the present diploma thesis is the Sentiment Analysis of the comments of the social media YouTube in cases of advertisement videos with the use of the Python programming language.
The Sentiment Analysis on a set of comments reveals the public’s reaction/response through commenting various videos and provides conclusions that could be utilized from the companies in order to improve the promotion strategy of their products.
In the present thesis, 78.534 comments are analysed – with the use of Emotion Lexicon – and classified based on the polarity (positive-negative) as well as based on the feeling that is expressed from the users. In parallel, the distribution of the positive and negative mood of the users is investigated per month and per year and conclusions are extracted concerning the reaction/response of the commenters to the advertisements.

Social, Political and Economic sciences

Sentiment analysis, marketing, social media, YouTube, comments, opinion mining

No

0

Yes

54

128

File access is restricted only to the intranet of UoA.

https://pergamos.lib.uoa.gr/uoa/dl/object/2820764