Anempirical Analysis of Classification Models for Detection of Fake News Articles

Telang, Hrishikesh; More, Shreya; Modi, Yatri; Kurup, Lakshmi

doi:10.1109/icecct.2019.8869504

Cited by 8 publications

(5 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…n-grams, tf-idf and Gradient Boosting Classifier, were conducted by Wynne and Wint [30]. Kaur et al and Telang et al evaluated multiple combinations of various classifier and extraction techniques in the fake news detection task [15,29]. Usage of the ensemble methods for disinformation detection was explored in [17].…”

Section: Related Workmentioning

confidence: 99%

Lifelong Learning Natural Language Processing Approach for Multilingual Data Classification

Kozal¹,

Leś²,

Zyblewski³

et al. 2022

Preprint

View full text Add to dashboard Cite

The abundance of information in digital media, which in today's world is the main source of knowledge about current events for the masses, makes it possible to spread disinformation on a larger scale than ever before. Consequently, there is a need to develop novel fake news detection approaches capable of adapting to changing factual contexts and generalizing previously or concurrently acquired knowledge. To deal with this problem, we propose a lifelong learning-inspired approach, which allows for fake news detection in multiple languages and the mutual transfer of knowledge acquired in each of them. Both classical feature extractors, such as Term frequency-inverse document frequency or Latent Dirichlet Allocation, and integrated deep nlp (Natural Language Processing) bert (Bidirectional Encoder Representations from Transformers) models paired with mlp (Multilayer Perceptron) classifier, were employed. The results of experiments conducted on two datasets dedicated to the fake news classification task (in English and Spanish, respectively), supported by statistical analysis, confirmed that utilization of additional languages could improve performance for traditional methods. Also, in some cases supplementing the deep learning method with classical ones can positively impact obtained results. The ability of models to generalize the knowledge acquired between the analyzed languages was also observed.

show abstract

Section: Related Workmentioning

confidence: 99%

Lifelong Learning Natural Language Processing Approach for Multilingual Data Classification

Kozal¹,

Leś²,

Zyblewski³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Tokenization aids in the splitting of phrases and stop words help in removing words that don't add much value and therefore makes the phrase or sentence more concise. It has also been noticed that neural networks trained with N-gram which assigns probabil-ities to sentences and sequences of words performed better in comparison to sequence vectors [2] [3]. Furthermore, Models that are trained on a dataset consisting of news content rather than headlines tend to achieve higher accuracy and require more computational time.…”

Section: A Literature Surveymentioning

confidence: 99%

Combating Misinformation Dissemination through Verification and Content Driven Recommendation

Hawa

Lobo

Dogra

et al. 2021

2021 Third International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV)

View full text Add to dashboard Cite

The COVID 19 pandemic is a humanitarian emergency that poses an enormous threat to society and has impacted various social media platforms and journalism. News and social media has become an immensely popular platform for consumption of information. However, these platforms are also the bearer of fake news and information which causes negative effects and creates panic. Thus, this research work aim to tackle this problem by creating a unique hybrid model using Machine learning algorithms with Natural Language Processing (NLP) techniques to verify news. In order to make the proposed system foolproof, a superior content based recommendation system is developed which will encourage users to consume authenticated news and content from verified sources. Thus, such a system will provide a holistic approach as it not only verifies but also provides genuine and true recommendations for the same.

show abstract

“…As mentioned in Section 3.2.4, mainstream datasets are relatively small for such a machine learning problem. This could be a reason as to why many authors combined datasets, in order to boost the number of articles for model training and to protect against issues such as overfitting, poor generalisability as well as bias (as justified in [61,62]). Existing annotation approaches, manual vs. automated annotation, have limitations which result in a trade-off; that is, datasets are: either well-labelled but smaller and expensive in terms of human labour, by using independent fact-checkers, or poorly labelled but larger using generic labelling algorithms, such as labelling based on the source domain name of the articles.…”

Section: Comparison Of Average Performance By Feature Group (Rq21)mentioning

confidence: 99%

A Systematic Review on the Detection of Fake News Articles

Hoy¹,

Koulouri²

2021

Preprint

View full text Add to dashboard Cite

It has been argued that fake news and the spread of false information pose a threat to societies throughout the world, from influencing the results of elections to hindering the efforts to manage the COVID-19 pandemic. To combat this threat, a number of Natural Language Processing (NLP) approaches have been developed. These leverage a number of datasets, feature extraction/selection techniques and machine learning (ML) algorithms to detect fake news before it spreads. While these methods are well-documented, there is less evidence regarding their efficacy in this domain. By systematically reviewing the literature, this paper aims to delineate the approaches for fake news detection that are most performant, identify limitations with existing approaches, and suggest ways these can be mitigated. The analysis of the results indicates that Ensemble Methods using a combination of news content and socially-based features are currently the most effective. Finally, it is proposed that future research should focus on developing approaches that address generalisability issues (which, in part, arise from limitations with current datasets), explainability and bias.

show abstract

Anempirical Analysis of Classification Models for Detection of Fake News Articles

Cited by 8 publications

References 2 publications

Lifelong Learning Natural Language Processing Approach for Multilingual Data Classification

Lifelong Learning Natural Language Processing Approach for Multilingual Data Classification

Combating Misinformation Dissemination through Verification and Content Driven Recommendation

A Systematic Review on the Detection of Fake News Articles

Contact Info

Product

Resources

About