Scholarlink

CommunicationFootball fake news detection between people has significantly improved recently because of the increase in online social mediaSocial media. People communicate, share information, and consume news through social networking platforms like Twitter, Facebook, WhatsApp, etc. Recent content shared on social mediaSocial media is frequently questionable and, in some cases, purposely misleading, and the term “fake news” is frequently used to describe this material. False news being widely disseminated online has the potential to be harmful to society. Fake news detection has become a crucial area of research that objective to identify whether a particular content is legitimate or fake. This paper presents an innovative approach for identifying fake information in football news that combines regularMachine learning machine learning (ML) and deep learning (DL)Deep learning approaches. The ML approach utilizes four models: Decision tree (DT)Decision Tree (DT), Random Forest (RF)Random Forest (RF), Support Vector Machine (SVM)Support Vector Machine (SVM), and Naive BayesNaive Bayes (NB) for training and evaluation. Different n-gram sizes are employed with TF-ID feature extraction to construct matrix features; also, grid search optimizes ML models through cross-validation. On the other hand, the DLDeep learning approach employs the TextRNNTextRNN model, which combines an Embedding layer and an RNN layer to do text classification tasks to evaluate the model. The TextRNNTextRNN model accurately classified fake football news and achieved the highest accuracy, with a recall of 94.50%, precision of 94.25%, and an outstanding F-score of 94.35%. It excelled in the fake news data and proved its superiority over other ML classifiers. In contrast, the ML models showed good performance, and the Random Forest (RF)Random Forest (RF) model acquired the best precision of 92.87%, recall of 92.10%, and F-score of 92.65%, among others. Due to the lack of study on football false news, it may be inferred from the experiment findings that the suggested method is superior to conventional methods.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Comparative Analysis of Machine Learning Classifiers for Football Fake News Detection: A TextRNN-Based Approach

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • V. Prema Manvi,
  • U. Saipriya,
  • Shaik Suhana,
  • S. Firoz Begum,
  • S. Abhitha Angel

摘要

CommunicationFootball fake news detection between people has significantly improved recently because of the increase in online social mediaSocial media. People communicate, share information, and consume news through social networking platforms like Twitter, Facebook, WhatsApp, etc. Recent content shared on social mediaSocial media is frequently questionable and, in some cases, purposely misleading, and the term “fake news” is frequently used to describe this material. False news being widely disseminated online has the potential to be harmful to society. Fake news detection has become a crucial area of research that objective to identify whether a particular content is legitimate or fake. This paper presents an innovative approach for identifying fake information in football news that combines regularMachine learning machine learning (ML) and deep learning (DL)Deep learning approaches. The ML approach utilizes four models: Decision tree (DT)Decision Tree (DT), Random Forest (RF)Random Forest (RF), Support Vector Machine (SVM)Support Vector Machine (SVM), and Naive BayesNaive Bayes (NB) for training and evaluation. Different n-gram sizes are employed with TF-ID feature extraction to construct matrix features; also, grid search optimizes ML models through cross-validation. On the other hand, the DLDeep learning approach employs the TextRNNTextRNN model, which combines an Embedding layer and an RNN layer to do text classification tasks to evaluate the model. The TextRNNTextRNN model accurately classified fake football news and achieved the highest accuracy, with a recall of 94.50%, precision of 94.25%, and an outstanding F-score of 94.35%. It excelled in the fake news data and proved its superiority over other ML classifiers. In contrast, the ML models showed good performance, and the Random Forest (RF)Random Forest (RF) model acquired the best precision of 92.87%, recall of 92.10%, and F-score of 92.65%, among others. Due to the lack of study on football false news, it may be inferred from the experiment findings that the suggested method is superior to conventional methods.