Naslov
Sentiment recognition based on an affective lexicon with continuous distribution of affective dimensions

Autori
Sanchez Gil, Alex

Vrsta, podvrsta i kategorija rada
Ocjenski radovi, diplomski rad, preddiplomski

Fakultet
Fakultet elektrotehnike i računarstva

Mjesto
Zagreb

Datum
12.07

Godina
2019

Stranica
41

Mentor
Bagić Babac, Marina

Ključne riječi
sentiment lexicon, social media, emotions

Sažetak
There are many publicly available affective lexicons for sentiment analysis. However, most of these lexicons contain words with discrete affective labels, and therefore, their applications are limited. The purpose of this thesis to design and implement an affective lexicon with continuous distribution of affective dimensions from a labelled dataset based on crowd intelligence. In addition, the algorithms for using this lexicon are also to be provided. A comparison of the results of sentiment analysis based on a set of lexicons and algorithms is to be elaborated.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo

POVEZANOST RADA