TwiSE at SemEval-2016 Task 4: Twitter Sentiment Classification

This paper describes SiTAKA, our system that has been used in task 4A, English and Arabic languages, Sentiment Analysis in Twitter of SemEval2017. The system proposes the representation of tweets using a novel set of features, which include a bag of negated words and the information provided by some lexicons. The polarity of tweets is determined by a classifier based on a Support Vector Machine. Our system ranks 2nd among 8 systems in the Arabic language tweets and ranks 8th among 38 systems in the English-language tweets.

In this paper we report our works for SemEval-2014 Sentiment Analysis in Twitter evaluation challenge. This is the first time we attempt for this task, and our submissions are based on supervised machine learning algorithm. We use Support Vector Machine for both the tasks, viz. contextual polarity disambiguation and message polarity classification. We identify and implement a small set of features for each the tasks, and did not make use of any external resources and/or tools. The systems are tuned on the development sets and finally blind evaluation is performed on the respective test set, which consists of the datasets of five different domains. Our submission for the first task shows the F-score values of 76.3%, 77.04%, 70.91%, 72.25% and 66.32% for LiveJournal2014, SMS2013, Twitter2013, Twitter2014 and Twitter2014Sarcasm datasets, respectively. The system developed for the second task yields the F-score values of 54.68%, 40.56%, 50.32%, 48.22% and 36.73%, respectively for the five different test datasets.

Twitter is attracting significant interests from the research community in the last few years. Sentiment analysis of tweets is among the hottest topics of research nowadays. State of the art approaches of sentiment analysis present many shortcomings when classifying tweets, in particular when the classification goes beyond the binary or ternary classification. Multi-class sentiment analysis has proven to be a very challenging task. This is mainly for the simple reason that a tweet usually does not contain a single sentiment, but many ones. In this paper, we propose a pattern-based approach for sentiment quantification in Twitter. By quantification, we refer to the detection of the existing sentiments within a tweet and the detection of the weight of these sentiments. In a first step, we classify tweets into positive, negative, or neutral. Our approach reaches an accuracy of 81%. We then perform the sentiment quantification on the sentimental tweets (i.e., positive and negative ones) to extract the sentiments within them: we define 5 positive sentiment sub-classes 5 negative ones and detect which exist in each tweet. We define 2 metrics to measure the correctness of sentiment detection, and prove that sentiment quantification can be a more meaningful task than the regular multi-class classification.

This paper describes our multi-view ensemble approach to SemEval-2017 Task 4 on Sentiment Analysis in Twitter, specifically, the Message Polarity Classification subtask for English (subtask A). Our system is a voting ensemble, where each base classifier is trained in a different feature space. The first space is a bag-of-words model and has a Linear SVM as base classifier. The second and third spaces are two different strategies of combining word embeddings to represent sentences and use a Linear SVM and a Logistic Regressor as base classifiers. The proposed system was ranked 18th out of 38 systems considering F1 score and 20th considering recall.

Sentiment Analysis as Reputational Risk Indicator

This paper describes our proposal for Sentiment Analysis in Twitter for the Spanish language. The main characteristics of the system are the use of word embedding specifically trained from tweets in Spanish and the use of self-attention mechanisms that allow to consider sequences without using convolutional nor recurrent layers. These self-attention mechanisms are based on the encoders of the Transformer model. The results obtained on the Task 1 of the TASS 2019 workshop, for all the Spanish variants proposed, support the correctness and adequacy of our proposal.

Sentiment analysis in Twitter has really engaged interest in field of research. Sentiment classification in Twitter deals with analyzing the tweets in terms of their sentiment polarity. The proposed method deals with twitter sentiment classification by employing a classification model of machine learning domain which makes use of different textual features viz. n-grams of twitter data. Also, we have used three different weighting schemes to understand the impact of weighting on classifier accuracy. Furthermore, a sentiment score vector of tweets is used to provide external knowledge in order to improve the performance of SVM classifier.

With the rapid growth of online social media content, and the impact these have made on people's behavior, many researchers have been interested in studying these media platforms. A major part of their work focused on sentiment analysis and opinion mining. These refer to the automatic identification of opinions of people toward specific topics by analyzing their posts and publications. Multi-class sentiment analysis, in particular, addresses the identification of the exact sentiment conveyed by the user rather than the overall sentiment polarity of his text message or post. That being the case, we introduce a task different from the conventional multi-class classification, which we run on a data set collected from Twitter. We refer to this task as ``quantification.”By the term ``quantification,”we mean the identification of all the existing sentiments within an online post (i.e., tweet) instead of attributing a single sentiment label to it. For this sake, we propose an approach that automatically attributes different scores to each sentiment in a tweet, and selects the sentiments with the highest scores which we judge as conveyed in the text. To reach this target, we added to our previously introduced tool SENTA the necessary components to run and perform such a task. Throughout this work, we present the added components; we study the feasibility of quantification, and propose an approach to perform it on a data set made of tweets for 11 different sentiment classes. The data set was manually labeled and the results of the automatic analysis were checked against the human annotation. Our experiments show the feasibility of this task and reach an F1 score equal to 45.9%.

We describe the systems we have used for participating in Subtasks D (binary quantification) and E (ordinal quantification) of SemEval-2016 Task 4 “Sentiment Analysis in Twitter”. The binary quantification system uses a “Probabilistic Classify and Count” (PCC) approach that leverages the calibrated probabilities obtained from the output of an SVM. The ordinal quantification approach uses an ordinal tree of PCC binary quantifiers, where the tree is generated via a splitting criterion that minimizes the ordinal quantification loss.

This paper describes the KLUEless system which participated in the SemEval-2015 task on “Sentiment Analysis in Twitter”. This year the updated system based on the developments for the same task in 2014 (Evert et al., 2014) and 2013 (Proisl et al., 2013) participated in all five subtasks. The paper gives an overview of the core features extended by different additional features and parameters required for individual subtasks. Experiments carried out after the evaluation period on the test dataset 2015 with the gold standard available are integrated into each subtask to explain the submitted feature selection.

In this paper we describe a Deep Convolutional Neural Network (DNN) approach to perform two sentiment detection tasks: message polarity classification and contextual polarity disambiguation.We apply the proposed approach for the SemEval-2014 Task 9: Sentiment Analysis in Twitter.Despite not using any handcrafted feature or sentiment lexicons, our system achieves very competitive results for Twitter data.

This paper presents the approach of the GTI Research Group to SemEval-2016 task 4 on Sentiment Analysis in Twitter, or more specifically, subtasks A (Message Polarity Classification), B (Tweet classification according to a two-point scale) and D (Tweet quantification according to a two-point scale). We followed a supervised approach based on the extraction of features by a dependency parsing-based approach using a sentiment lexicon and Natural Language Processing techniques.

We present, in this paper, our contribution in SemEval2017 task 4 : Sentiment Analysis in Twitter , subtask A: Message Polarity Classification , for En-glish and Arabic languages. Our system is based on a list of sentiment seed words adapted for tweets. The sentiment relations between seed words and other terms are captured by cosine similarity between the word embedding representations (word2vec). These seed words are extracted from datasets of annotated tweets available online. Our tests, using these seed words, show significant improvement in results compared to the use of Turney and Littman's (2003) seed words, on polarity classification of tweet messages.

This paper describes the system we submitted to the SemEval-2014 shared task on sentiment analysis in Twitter. Our system is a hybrid combination of two system developed for a course project at CMUQatar. We use an SVM classifier and couple a set of features from one system with feature and parameter optimization framework from the second system. Most of the tuning and feature selection efforts were originally aimed at task-A of the shared task. We achieve an F-score of 84.4% for task-A and 62.71% for task-B and the systems are ranked 3rd and 29th respectively.

This paper describes the system that was submitted to SemEval2015 Task 10: Sentiment Analysis in Twitter. We participated in Subtask B: Message Polarity Classification. The task is a message level classification of tweets into positive, negative and neutral sentiments. Our model is primarily a supervised one which consists of well designed features fed into an SVM classifier. In previous runs of this task, it was found that lexicons played an important role in determining the sentiment of a tweet. We use existing lexicons to extract lexicon specific features. The lexicon based features are further augmented by tweet specific features. We also improve our system by using acronym and emoticon dictionaries. The proposed system achieves an F1 score of 59:83 and 67:04 on the Test Data and Progress Data respectively. This placed us at the 18 th position for the Test Dataset and the 16 th position for the Progress Test Dataset.