Deep Ensemble Model with the Fusion of Character, Word and Lexicon Level Information for Emotion and Sentiment Prediction

Ghosal, Deepanway; Akhtar, Md. Shad; Ekbal, Asif; Bhattacharyya, Pushpak

doi:10.1007/978-3-030-04221-9_15

Cited by 6 publications

(8 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…various state-of-the-art and traditional techniques; and e) We also presented a detailed qualitative analysis on the errors encountered by our proposed method. In another work, Ghosal et al [18] developed a deep ensemble model that utilizes the character, word and lexicon level fusion for the sentiment and emotion prediction. The main contributions of our proposed work are highlighted below: a) We effectively combine deep learning and feature driven traditional model via an ensemble framework; b) We develop a stacked denoising autoencoder based technique for an enhanced word representation by leveraging the syntactic and semantic richness of the two distributed word representations; c) We perform normalization of tweets by utilizing various heuristics; and d) We build a stateof-the-art model that effectively solves both the problems of emotion analysis and sentiment analysis.…”

Section: Introductionmentioning

confidence: 99%

How Intense Are You? Predicting Intensities of Emotions and Sentiments using Stacked Ensemble [Application Notes]

Akhtar

Ekbal

Wang

2020

IEEE Comput. Intell. Mag.

Self Cite

219

View full text Add to dashboard Cite

E motions and sentiments are subjective in nature. They differ on a case-to-case basis. However, predicting only the emotion and sentiment does not always convey complete information. The degree or level of emotions and sentiments often plays a crucial role in understanding the exact feeling within a single class (e.g., 'good' versus 'awesome'). In this paper, we propose a stacked ensemble method for predicting the degree of intensity for emotion and sentiment by combining the outputs obtained from several deep learning and classical feature-based models using a multi-layer perceptron network. We develop three deep learning models based on convolutional neural network, long short-term memory and gated recurrent unit and one classical supervised model based on support vector regression. We evaluate our proposed technique for two problems, i.e., emotion analysis in the generic domain and sentiment analysis in the financial domain. The proposed model shows impressive results for both the problems. Comparisons show that our proposed model achieves improved performance over the existing state-of-theart systems.

show abstract

Section: Introductionmentioning

confidence: 99%

How Intense Are You? Predicting Intensities of Emotions and Sentiments using Stacked Ensemble [Application Notes]

Akhtar

Ekbal

Wang

2020

IEEE Comput. Intell. Mag.

Self Cite

219

View full text Add to dashboard Cite

show abstract

“…The typical tasks to which sentiment analysis and affective computing are jointly applied, are emotion recognition and polarity detection [22,[47][48][49][50], and different approaches can be found, including semantic, statistical and hybrid approaches. Semantic approaches [51] are typically based on the use of lexicons such as Affective Lexicon [2], WordNet-Affect [3], Senti-WordNet [4], EmotiNet [52], SenticNet [7] or lexicons based on Russell's dimensions [53,54], among others.…”

Section: Affective Computing and Sentiment Analysismentioning

confidence: 99%

Ordered Weighted Averaging for Emotion-Driven Polarity Detection

2021

View full text Add to dashboard Cite

The overall rating of an opinion can generally be considered as the aggregation of the individual ratings of all features of that opinion. Nevertheless, there are cases in which the overall rating differs substantially from the mean or weighted mean of the ratings of the individual features. These cases can be explained in terms of user mood. To address this problem, this study introduces a fuzzy framework for computing user mood based on SenticNet and sentic patterns, which are used to guide an ordered weighted averaging operator. This operator allows the aggregation to be computed in such a way as to provide an understanding of why some positive or negative aspects are considered to a greater or lesser extent. The performance and advantages of this proposal are illustrated in depth via a variety of scenarios applied to real data. The results show a promising framework applicable to other tools, such as customized recommender systems or decision support systems.

show abstract

“…Further to the quoted algorithms, 22 studies [317,452,505,319,124,512,461,488,489,344,448,352,353,356,47,449,381,386,387,275,475,117] used ensemble learning methods in their work, where they combined the output of several base machine learning and/or deep learning methods. In particular, [117] compared eight popular lexicon and machine learning based sentiment analysis algorithms, and then developed an ensemble that combines them, which in turn provided the best coverage results and competitive agreement.…”

Section: Hybrid (Hy)mentioning

confidence: 99%

“…In particular, [117] compared eight popular lexicon and machine learning based sentiment analysis algorithms, and then developed an ensemble that combines them, which in turn provided the best coverage results and competitive agreement. Moreover, [505] proposes an MLP-based ensemble network that combines LSTM, CNN and feature-based MLP models, with each model incorporating character, word and lexicon level information, to predict the degree of intensity for sentiment and emotion. Lastly, as presented in Table 11, the RF ensemble learning method was used in the 21 studies [275,391,513,337,384,356,357,330,349,344,292,512,341,491,453,486,48,319,477,285,195].…”

Section: Hybrid (Hy)mentioning

confidence: 99%

Over a Decade of Social Opinion Mining: A Systematic Review

Cortis,

Davis

2020

Preprint

View full text Add to dashboard Cite

Social media popularity and importance is on the increase, due to people using it for various types of social interaction across multiple channels. This social interaction by online users includes submission of feedback, opinions and recommendations about various individuals, entities, topics, and events. This systematic review focuses on the evolving research area of Social Opinion Mining, tasked with the identification of multiple opinion dimensions, such as subjectivity, sentiment polarity, emotion, affect, sarcasm and irony, from user-generated content represented across multiple social media platforms and in various media formats, like text, image, video and audio. Therefore, through Social Opinion Mining, natural language can be understood in terms of the different opinion dimensions, as expressed by humans. This contributes towards the evolution of Artificial Intelligence, which in turn helps the advancement of several real-world use cases, such as customer service and decision making. A thorough systematic review was carried out on Social Opinion Mining research which totals 485 studies and spans a period of twelve years between 2007 and 2018. The in-depth analysis focuses on the social media platforms, techniques, social datasets, language, modality, tools and technologies, natural language processing and other

show abstract

Deep Ensemble Model with the Fusion of Character, Word and Lexicon Level Information for Emotion and Sentiment Prediction

Cited by 6 publications

References 18 publications

How Intense Are You? Predicting Intensities of Emotions and Sentiments using Stacked Ensemble [Application Notes]

How Intense Are You? Predicting Intensities of Emotions and Sentiments using Stacked Ensemble [Application Notes]

Ordered Weighted Averaging for Emotion-Driven Polarity Detection

Over a Decade of Social Opinion Mining: A Systematic Review

Contact Info

Product

Resources

About