SentiGAN: Generating Sentimental Texts via Mixture Adversarial Networks

By Ke Wang, Xiaojun Wan. 2018

198 citations recorded by the source corpus.

Generating texts of different sentiment labels is getting more and more attention in the area of natural language generation. Recently, Generative Adversarial Net (GAN) has shown promising results in text generation. However, the texts generated by GAN usually suffer from the problems of poor quality, lack of diversity and mode collapse. In this paper, we propose a novel framework - SentiGAN, which has multiple generators and one multi-class discriminator, to address the above problems. In our framework, multiple generators are trained simultaneously, aiming at generating texts of different sentiment labels without supervision. We propose a penalty based objective in the generators to force each of them to generate diversified examples of a specific sentiment label. Moreover, the use of multiple generators and one multi-class discriminator can make each generator focus on generating its own examples of a specific sentiment label accurately. Experimental results on four datasets demonstrate that our model consistently outperforms several state-of-the-art text generation methods in the sentiment accuracy and quality of generated texts.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

GAN Memory with No Forgetting (2020, background): “Generative adversarial networks (GANs) GANs continue to show increasing power in synthesizing highly realistic observations [32, 45, 51, 9, 33, 34], and have found wide applicability in various ﬁelds [39, 1, 19, 79, 81, 82, 12, 87, 37].”
Challenges in Building Intelligent Open-domain Dialog Systems (2019, background): “In [98], to generate reviews of a particular polarity, the authors proposed a multi-class generative adversarial network which consists of multiple generators for multi-class polarities and a multi-class discriminator.”
Human-Machine Collaboration Approaches to Build a Dialogue Dataset for Hate Speech Countering (2022, methodology): “Novelty is utilized to check the quality of a generated dialogue by measuring its lexical difference with respect to a reference set of dialogues, and it is grounded on Jaccard similarity (Dziri et al., 2019; Wang and Wan, 2018).”
On Leveraging Pretrained GANs for Generation with Limited Data (2020, background): “the high-level semantically specific filters are also transferred, the similarity between the source and target domains is often critical for a beneficial transfer (Wang et al., 2018b). Similarly, based on the assumption that the source and target domains share the same support, Wang et al. (2020) i”
I Know the Feeling: Learning to Converse with Empathy (2018, background): “…works focus on controlling the emotional content of a text response either through a manually speciﬁed target (Zhou & Wang, 2018; Zhou et al., 2017; Wang & Wan, 2018; Hu et al., 2017; Huang et al., 2018) or through a general term to encourage higher levels of affect (Asghar et al., 2018), with…”
Beneath the Tip of the Iceberg: Current Challenges and New Directions in Sentiment Analysis Research (2020, background): “[75], GANs, Wang &Wan [217] or Seq2Seqmodels, Radford et al.”
A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications (2020, methodology): “SentiGAN [212] uses a mixture of generators and a multi-class discriminator to generate sentimental texts.”
Towards Empathetic Open-domain Conversation Models: A New Benchmark and Dataset (2018, background): “…focused on controlling the emotional content of a text response either through a manually speciﬁed target (Zhou and Wang, 2018; Zhou et al., 2018; Wang and Wan, 2018; Hu et al., 2017; Huang et al., 2018) or through a general term to encourage higher levels of affect (Asghar et al., 2018), with…”
Open Event Extraction from Online Text using a Generative Adversarial Network (2019, background): “Along this line, Wang et al. (2018) proposed SentiGAN to generate texts of different sentiment labels.”
HappyBot: Generating Empathetic Dialogue Responses by Improving User Experience Look-ahead (2019, background): “Tackling this problem, several recent works have been successful in controlling and conditioning the generated responses to certain sentiments, emotions, and emojis (Hu et al., 2017; Wang and Wan, 2018; Zhou and Wang, 2018; Zhou et al., 2018).”
LARA: Attribute-to-feature Adversarial Learning for New-item Recommendation (2020, background): “Moreover, it has already gained huge success in many tasks, such as image generation [22, 40], natural language generation [15, 34, 41], and music generation [9, 10].”
A Survey of Controllable Text Generation Using Transformer-based Pre-trained Language Models (2022, methodology): “After that, there is a surge of methods based on deep generative models, such as Variational Autoencoders (VAEs) [48, 125, 138, 142, 149, 154], Generative Adversarial Networks (GANs) [117, 140], and Energy-based Models [8, 25, 135, 166].”
On Leveraging Pretrained GANs for Limited-Data Generation (2020, background): “…to various research fields, such as image (Hoffman et al., 2017; Ledig et al., 2017; Ak et al., 2019), text (Lin et al., 2017; Fedus et al., 2018; Wang & Wan, 2018; Wang et al., 2019), video (Mathieu et al., 2015; Wang et al., 2017; 2018a; Chan et al., 2019), and audio (Engel et al., 2019;…”