SemEval-2016 Task 7: Determining Sentiment Intensity of English and Arabic Phrases

Abstract

We present a shared task on automatically determining sentiment intensity of a word or a phrase. The words and phrases are taken from three domains: general English, English Twitter, and Arabic Twitter. The phrases include those composed of negators, modals, and degree adverbs as well as phrases formed by words with opposing polarities. For each of the three domains, we assembled the datasets that include multi-word phrases and their constituent words, both manually annotated for real-valued sentiment intensity scores. The three datasets were presented as the test sets for three separate tasks (each focusing on a specific domain). Five teams submitted nine system outputs for the three tasks. All datasets created for this shared task are freely available to the research community.

Optimization of sentiment analysis using machine learning classifiers

“…This availability of data entices the interest of young researchers to plunge them in the field of sentiment analysis. People express their emotions and perspectives on the social media discussion forums [6]. The business organizations employ researchers to investigate the unrevealed facts about their products and services.…”

“…The dichotomy of sentiment is generally decided by the mindset of an author of text whether he is positively or negatively oriented towards his saying [6,[11][12][13]. Naïve Bayes classifier is a popular supervised classifier, furnishes a way to express positive, negative and neutral feelings in the web text.…”

Semantic Sentiment Analysis of Twitter Data

“…Other related tasks have explored aspect-based sentiment analysis [49,48,50], sentiment analysis of figurative language on Twitter [17], implicit event polarity [57], stance in tweets [37], out-of-context sentiment intensity of phrases [24], and emotion detection [66]. Some of these tasks featured languages other than English.…”

“…The importance of building sentiment polarity lexicons has resulted in a special subtask [55] at SemEval-2015 (part of Task 4) and an entire task [24] at SemEval-2016 (namely, Task 7), on predicting the out-of-context sentiment intensity of words and phrases. Yet, we should note though that the utility of using sentiment polarity lexicons for sentiment analysis probably needs to be revisited, as the best system at SemEval-2016 Task 4 could win without using any lexicons [12]; it relied on semi-supervised learning using a deep neural network.…”

Feature-Enriched Character-Level Convolutions for Text Regression

“…Unlike typical Sentiment Analysis tasks, which are set up as either binary or multiclass classification problems that require one to determine the opinion or sentiment in a given text, EIA aims at quantifying a certain emotion in a text, such as fear, anger, joy, sadness, etc. In the Emotion Intensity shared task of SemEval 2016 (Kiritchenko et al, 2016), which is the first of its kind, none of the five systems submitted employ neural regressors. We were also unable to find any other contributions outside the SemEval 2016 task that explore neural approaches to EIA.…”