Learning Textual Representations from Multiple Modalities to Detect Fake News Through One-Class Learning

Gôlo, Marcos Paulo Silva; Caravanti, Mariana; Rossi, Rafael; Rezende, Solange Oliveira; Nogueira, Bruno; Marcacini, Ricardo Marcondes

doi:10.1145/3470482.3479634

Cited by 10 publications

(8 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One-Class Learning train with only one class and predict examples as belonging to the interest class or not [Emmert-Streib and Dehmer 2022]. We can define OCL as [Gôlo et al 2021]:…”

Section: One-class Learningmentioning

confidence: 99%

Unsupervised Heterogeneous Graph Neural Network for Hit Song Prediction through One Class Learning

Silva

Gôlo

Marcacini

2022

Anais Do X Symposium on Knowledge Discovery, Mining and Learning (KDMiLe 2022)

View full text Add to dashboard Cite

Although the concept of success is subjective, it can be related to the popularity and interest of users. Measuring the success of a song in advance allows for offering information of great interest to the music market. Hit song prediction is an existing task in Music Information Retrieval that explores approaches for measuring music success based on features. Musical data is intrinsically multimodal, where features from different sources have complementary semantic information. Therefore, structuring musical data and building a unique space that embeds multiple features is a challenge in musical data representation. Using heterogeneous graphs to structure multimodal data is a resource for exploring the intrinsic semantic relationship between features. In this sense, this work proposes to structure musical features over heterogeneous graphs and learn a new graph-based multimodal representation for songs using an unsupervised graph neural network to handle the hit song prediction task. We formulated the hit song prediction task as a one-class learning problem to mitigate the non-hit song gaps and highlight the hit song as the interest class. We measure the performance of representations based on lyrics and artist features and present promising results using our learned representations that outperform other strategies for representing musical data.

show abstract

“…One-Class Learning train with only one class and predict examples as belonging to the interest class or not [Emmert-Streib and Dehmer 2022]. We can define OCL as [Gôlo et al 2021]:…”

Section: One-class Learningmentioning

confidence: 99%

Unsupervised Heterogeneous Graph Neural Network for Hit Song Prediction through One Class Learning

Silva

Gôlo

Marcacini

2022

Anais Do X Symposium on Knowledge Discovery, Mining and Learning (KDMiLe 2022)

View full text Add to dashboard Cite

show abstract

“…OCL uses only examples from one class (interest class) to learn, i.e., the learning is in the absence of counterexamples [Tax 2001]. OCL will be able to identify whether an instance belongs to the interest class, reducing the labeling effort and being more appropriate for open-domain applications or applications in which the user is interested in one class [Gôlo et al 2021a, Gôlo et al 2021b].…”

Section: Introductionmentioning

confidence: 99%

“…Second, The study applies MVAE in three real scenarios. First, the study detects fake news through OCL with the MVAE representations [Gôlo et al 2021a]. The MVAE learns a new representation from the combination of promising modalities: text embeddings, linguistic features, and topic/density information.…”

Section: Introductionmentioning

confidence: 99%

Text Representation through Multimodal Variational Autoencoder for One-Class Learning

Gôlo

Marcacini

2023

Anais Do XXXVI Concurso De Teses E Dissertações (CTD 2023)

View full text Add to dashboard Cite

Multi-class learning (MCL) methods perform Automatic Text Classification (ATC), which requires labeling for all classes. MCL fails when there is no well-defined information about the classes and requires a great effort to label instances. One-Class Learning (OCL) can mitigate these limitations since the training only has instances from one class, reducing the labeling effort and making the ATC more appropriate for open-domain applications. However, OCL is more challenging due to the lack of counterexamples for model training, requiring more robust representations. However, most studies use unimodal representations, even though different domains contain other information that can be used as modalities. Thus, this study proposes the Multimodal Variational Autoencoder (MVAE) for OCL. MVAE is a multimodal method that learns a new representation from more than one modality, capturing the characteristics of the interest class in an adequate way. MVAE explores semantic, density, linguistic, and spatial information modalities. The main contributions are: (i) a multimodal method for ATC through OCL; (ii) MVAE for fake news detection; (iii) relevant reviews detection via MVAE; and (iv) sensing events through MVAE.

show abstract

“…OCL uses only instances from one class to learn [22]. OCL will be able to identify whether an instance belongs to the interest class, reducing the labeling effort and being more appropriate for open-domain applications or one-class applications [6,10].…”

Section: Introductionmentioning

confidence: 99%

“…Even so, studies use the traditional Bag-of-Words (BoW) technique [12,16]. Other studies explore dimensionality reduction techniques [8,14]. Finally, studies used language models via neural networks [17,19].…”

Section: Introductionmentioning

confidence: 99%

Text Representation through Multimodal Variational Autoencoder for One-Class Learning

Gôlo,

Marcacini

2023

Anais Estendidos Do XXIX Simpósio Brasileiro De Sistemas Multimídia E Web (WebMedia 2023)

View full text Add to dashboard Cite

Multi-class learning (MCL) methods perform Automatic Text Classification (ATC), which requires labeling for all classes. MCL fails when there is no well-defined class information and requires a great eff ort in labeling. One-Class Learning (OCL) can mitigate these limitations since the training only has instances from one class, reducing the labeling eff ort and making the ATC more appropriate for open-domain applications. However, OCL is more challenging due to the lack of counterexamples. Even so, most studies use unimodal representations, even though different domains contain other information (modalities). Thus, this study proposes the Multimodal Variational Autoencoder (MVAE) for OCL. MVAE is a multimodal method that learns a new representation from more than one modality, capturing the characteristics of the interest class in an adequate way. MVAE explores semantic, density, linguistic, and spatial information modalities. The main contribution is a new multimodal method for representation learning on OCL scenarios considering few instances to train with state-of-the-art results in three domains.

show abstract

Learning Textual Representations from Multiple Modalities to Detect Fake News Through One-Class Learning

Cited by 10 publications

References 23 publications

Unsupervised Heterogeneous Graph Neural Network for Hit Song Prediction through One Class Learning

Unsupervised Heterogeneous Graph Neural Network for Hit Song Prediction through One Class Learning

Text Representation through Multimodal Variational Autoencoder for One-Class Learning

Text Representation through Multimodal Variational Autoencoder for One-Class Learning

Contact Info

Product

Resources

About