Naslov
Sentiment Analysis of Movie Reviews

Autori
Vrgoč, Katarina Anđela ; Pauković, Matija ; Russo, Mladen ; Šarić, Matko

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), stručni

Izvornik
26th International Conference on Software, Telecommunications and Computer Networks (SoftCOM 2018) : proceedings / - , 2018, 1-6

Skup
International Conference on Software, Telecommunications and Computer Networks (26 ; 2018)

Mjesto i datum
Supetar, Hrvatska, 13.09.2018. - 15.09.2018

Vrsta sudjelovanja
Radionica

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
sentiment analysis, opinion mining, n-grams, weight method

Sažetak
This paper addresses the problem of sentiment analysis and text sentiment prediction using a weight-oriented method applied on the dataset consisted of movie reviews provided within Kaggle Bag of Words Meets Bags of Popcorns competition. The method utilizes traditional Bag of Words model for vocabulary synthesis and further applies weight-oriented classifier for sentiment prediction. Finally, as different sets of N-grams i.e. unigrams, bigrams and trigrams were used, sentiment prediction accuracy is presented, compared and discussed for optimal approach selection

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo, Informacijske i komunikacijske znanosti

POVEZANOST RADA