Multiclass Classification for Bangla News Tags with Parallel CNN Using Word Level Data Augmentation

Amin, Ruhul; Sworna, Nabila Sabrin; Hossain, Nahid

doi:10.1109/tensymp50017.2020.9230981

Cited by 9 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Text-GCN was found to have superior accuracy, precision, recall, and F1 score compared to the other models. Despite data scarcity, the study demonstrates Text-GCN's promise for categorizing Bangla news items [21,22]. Another example of text classification for depression intensity detection can be seen in [23].…”

Section: Related Workmentioning

confidence: 80%

Unifying Sentence Transformer Embedding and Softmax Voting Ensemble for Accurate News Category Prediction

2023

View full text Add to dashboard Cite

The study focuses on news category prediction and investigates the performance of sentence embedding of four transformer models (BERT, RoBERTa, MPNet, and T5) and their variants as feature vectors when combined with Softmax and Random Forest using two accessible news datasets from Kaggle. The data are stratified into train and test sets to ensure equal representation of each category. Word embeddings are generated using transformer models, with the last hidden layer selected as the embedding. Mean pooling calculates a single vector representation called sentence embedding, capturing the overall meaning of the news article. The performance of Softmax and Random Forest, as well as the soft voting of both, is evaluated using evaluation measures such as accuracy, F1 score, precision, and recall. The study also contributes by evaluating the performance of Softmax and Random Forest individually. The macro-average F1 score is calculated to compare the performance of different transformer embeddings in the same experimental settings. The experiments reveal that MPNet versions v1 and v3 achieve the highest F1 score of 97.7% when combined with Random Forest, while T5 Large embedding achieves the highest F1 score of 98.2% when used with Softmax regression. MPNet v1 performs exceptionally well when used in the voting classifier, obtaining an impressive F1 score of 98.6%. In conclusion, the experiments validate the superiority of certain transformer models, such as MPNet v1, MPNet v3, and DistilRoBERTa, when used to calculate sentence embeddings within the Random Forest framework. The results also highlight the promising performance of T5 Large and RoBERTa Large in voting of Softmax regression and Random Forest. The voting classifier, employing transformer embeddings and ensemble learning techniques, consistently outperforms other baselines and individual algorithms. These findings emphasize the effectiveness of the voting classifier with transformer embeddings in achieving accurate and reliable predictions for news category classification tasks.

show abstract

Section: Related Workmentioning

confidence: 80%

Unifying Sentence Transformer Embedding and Softmax Voting Ensemble for Accurate News Category Prediction

2023

View full text Add to dashboard Cite

show abstract

“…The existing paper in different research sectors basically works on dataset of Bengali comments, tweets, magazines or political, sports news. Bengali comments, text, etc., are quite unofficial data as far need much moderation [12]. Again, for classifying political news, problem will arise dealing with limited terms [13].…”

Section: Resultsmentioning

confidence: 99%

Bangla News Classification Using GloVe Vectorization, LSTM, and CNN

Chowdhury

Eumi

Sarkar

et al. 2021

Lecture Notes on Data Engineering and Communications Technologies

View full text Add to dashboard Cite

Text mining has gained considerable popularity over the last few years. Since the awareness of mobile media, streaming media, print media, and many other outlets are now accessible to consumers. Due to the large availability of text in many respects, research experts registered many unstructured data and found various ways in the literature of converting this dispersed text into a given, organized volume. Compared to the short text, the emphasis on complete classification (complete news, big records, long text, etc.) is prevalent. We addressed in this paper the process of text classification, grading, and various methodologies for feature extraction in short texts, i.e., news classification based on their headlines. Existing classification is compared and their operating methodologies presented efficiently. This work serves the purpose of classifying different types of Bangla Newspaper articles into 10 specific categories. The classification task is performed on Bengali text from three renowned newspapers of Kolkata. We have used advanced data tokenization techniques and unsupervised 'GloVe' vectorization for better classification performance. We applied LSTM and CNN as our main feature extractors. Comparing with other models like binary SVM classifier, standard LSTM, BiLSTM, CNN, or ANN, this proposed work gives better accuracy of 87%.

show abstract

Deep Learning Based Software Vulnerability Detection in Code Snippets and Tag Questions Using Convolutional Neural Networks

Khanra,

Krishna,

Jeevan Samrudh

et al. 2024

Communications in Computer and Information Science

View full text Add to dashboard Cite

Multiclass Classification for Bangla News Tags with Parallel CNN Using Word Level Data Augmentation

Cited by 9 publications

References 4 publications

Unifying Sentence Transformer Embedding and Softmax Voting Ensemble for Accurate News Category Prediction

Unifying Sentence Transformer Embedding and Softmax Voting Ensemble for Accurate News Category Prediction

Bangla News Classification Using GloVe Vectorization, LSTM, and CNN

Deep Learning Based Software Vulnerability Detection in Code Snippets and Tag Questions Using Convolutional Neural Networks

Contact Info

Product

Resources

About