Madisetty, Sreekanth and Desarkar, Maunendra Sankar
(2017)
NSEmo at EmoInt-2017: An Ensemble to Predict Emotion Intensity in Tweets.
In: 8th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis, 7–11, September 2017, Copenhagen, Denmark.
Abstract
In this paper, we describe a method to pre-
dict emotion intensity in tweets. Our ap-
proach is an ensemble of three regression
methods. The first method uses content-
based features (hashtags, emoticons, elon-
gated words, etc.). The second method
considers word n-grams and character n-
grams for training.
The final method
uses lexicons, word embeddings, word n-
grams, character n-grams for training the
model. An ensemble of these three meth-
ods gives better performance than individ-
ual methods. We applied our method on
WASSA emotion dataset. Achieved re-
sults are as follows: average Pearson cor-
relation is 0.706, average Spearman cor-
relation is 0.696, average Pearson corre-
lation for gold scores in range 0.5 to 1 is
0.539, and average Spearman correlation
for gold scores in range 0.5 to 1 is 0.514.
Actions (login required)
 |
View Item |
Altmetric
Altmetric