Sentiment Lexicon-Based Features for Sentiment Analysis in Short Text

Hamdan, Hussam; Bellot, Patrice; Béchet, Frédéric

doi:10.13053/rcs-90-1-17

Cited by 17 publications

(21 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Higher scores (bold) are the ones resulting from sentiment-related features: the mood and the polarities from Sen-tiStrength [21]. The average recall is even higher adding Echo [5] Accuracy Precision Recall F1-score Features Lemmas*, RandomForest importance scores ranked the most discriminative features: mood came first in every run, followed by Echo Neutral Label. However, these scores do not take into account the combination of multiple features (i.e.…”

Section: Results Analysismentioning

confidence: 99%

“…As textual features, we used bags of words or bags of characters, total word count, exclamation and interrogation marks and n-grams (up to 5-grams). As sentiment-related features, we used positive, negative, and neutral polarity scores from SentiStrength 8 [21], using the available model trained on MySpace comments and tweets, and from Echo 9 [5] trained on 9684 tweets. Another sentiment-related feature is the current mood selected by the user: users can choose between 38 moods and change it whenever they want.…”

Section: Emoji Prediction 31 Data Analysismentioning

confidence: 99%

See 1 more Smart Citation

Emoji recommendation in private instant messages

Guibon¹,

Ochs

Bellot

2018

Proceedings of the 33rd Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

Emojis are some of the most common ways to convey emotions and sentiments in social messaging applications. In order to help the user choose emojis among a vast range of possibilities, we aim at developing an automatic recommendation system based on user message analysis and real emoji usage, which goes beyond the simple dictionnary lookup that is done in the industry (mainly Android and iOS). For this purpose, we present a novel automatic emoji prediction model trained and tested on real data and based on sentiment-related features. Such a model differ from the ones learnt from tweets and can predict emojis with a 84.48% f1-score and a 95.49% high precision, using MultiLabel-RandomForest algorithm on real private instant message corpus. We want to determine the best discriminative features for this task.

show abstract

Section: Results Analysismentioning

confidence: 99%

Section: Emoji Prediction 31 Data Analysismentioning

confidence: 99%

Emoji recommendation in private instant messages

Guibon¹,

Ochs

Bellot

2018

Proceedings of the 33rd Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Future perspective of this work will focus on improving the underlying components of our architecture. A deeper analysis of the available unstructured content would make it possible to infer the context, the mentioned named entities in items' descriptions or users' sentiments and needs expressed in their comments and queries (Hamdan et al, 2015). Such analysis would make items' clustering and users interests inference more accurate and thus increasing the quality of the recommendations.…”

Section: Discussionmentioning

confidence: 99%

An adaptive and interactive recommendation model for consumers' behaviours prediction

Haddad

Baazaoui

2018

IJDMMM

View full text Add to dashboard Cite

Recommendation algorithms aim at predicting customers' interests and purchases using different ideas and hypotheses. Consequently, system designers need to choose the recommendation approach that is the most suitable with regard to their products' nature and consumers' behaviours within the application field. In this paper, we propose an adaptive recommendation model based on statistical modelling to assist consumers facing choice overload by predicting their interests and consumption behaviours. We also propose a dynamic variant of the model taking into account the recommendations' time-value during interactive online recommendation scenarios. Our proposal has endured a two-fold evaluation. First, we conducted an offline comparative study on the MovieLens recommendation dataset in order to assess our model's performance with regard to several widely adopted recommendation techniques. Then, the model was evaluated within a real time online news recommendation platform to highlight its adaptability, scalability and efficiency in a highly interactive application domain.

show abstract

“…In sentiment analysis, the lexicon-based approach is also used, which relies on sentiment lexicons having positive, negative, and neutral terms or tones. The sentiment detection based on lexicon-based properties depends upon the co-occurrence of a set of words contained by a sentiment lexicon [27,33].…”

Section: Methodsmentioning

confidence: 99%

Sentiment analysis on IMDB using lexicon and neural networks

et al. 2020

View full text Add to dashboard Cite

To find out what other people think has been an essential part of information-gathering behaviors. And in the case of movies, the movie reviews can provide an intricate insight into the movie and can help decide whether it is worth spending time on. However, with the growing amount of data in reviews, it is quite prudent to automate the process, saving on time. Sentiment analysis is an important field of study in machine learning that focuses on extracting information of subject from the textual reviews. The area of analysis of sentiments is related closely to natural language processing and text mining. It can successfully be used to determine the attitude of the reviewer in regard to various topics or the overall polarity of the review. In the case of movie reviews, along with giving a rating in numeric to a movie, they can enlighten us on the favorableness or the opposite of a movie quantitatively; a collection of those then gives us a comprehensive qualitative insight on different facets of the movie. Opinion mining from movie reviews can be challenging due to the fact that human language is rather complex, leading to situations where a positive word has a negative connotation and vice versa. In this study, the task of opinion mining from movie reviews has been achieved with the use of neural networks trained on the "Movie Review Database" issued by Stanford, in conjunction with two big lists of positive and negative words. The trained network managed to achieve a final accuracy of 91%.

show abstract

Sentiment Lexicon-Based Features for Sentiment Analysis in Short Text

Cited by 17 publications

References 12 publications

Emoji recommendation in private instant messages

Emoji recommendation in private instant messages

An adaptive and interactive recommendation model for consumers' behaviours prediction

Sentiment analysis on IMDB using lexicon and neural networks

Contact Info

Product

Resources

About