Hi,
My field of research is the multi-label classification of texts. I would be grateful if you have any information on this subject.
In this post, the author builds a multi-label model that is capable of detecting different types of toxic comments like severe toxic, threats, obscenity, insults, and so on, by using OneVsRest multi-label strategy apllied to a Kaggle challenge dataset: Toxic Comment Classification Challenge.
https://towardsdatascience.com/multi-label-text-classification-with-scikit-learn-30714b7819c5
Dear Houra Zahed ,
Plus more resources for you
https://towardsdatascience.com/building-a-multi-label-text-classifier-using-bert-and-tensorflow-f188e0ecdc5d
https://medium.com/huggingface/multi-label-text-classification-using-bert-the-mighty-transformer-69714fa3fb3d
https://www.analyticsvidhya.com/blog/2019/04/predicting-movie-genres-nlp-multi-label-classification/
And this is what I like the most, you can check the state-of-the-art model and code in this page:
https://paperswithcode.com/task/multi-label-text-classification
Here are some good github repos for multilabel text classification which might be useful for you:
1. https://github.com/RandolphVI/Multi-Label-Text-Classification
2. https://github.com/ammarinjtk/Multi-Label-Text-Classification
3. https://github.com/chenzhi1992/Multi-Label-Text-Classification
4. https://github.com/lonePatient/Bert-Multi-Label-Text-Classification
yes you can use a bag of word model (SVM, Gradient Boosting, Nearest Neighbor, Naive Bayes) or if you want to take word order into account, a string or tree kernel (latter usually based on Dependency Trees). For multi label categorization with SVM you usually use the methods "1 against all" or "all against all".
Hello Houra Zahed this article might be helpful if you are looking for non-deeplearning approaches :
1. https://towardsdatascience.com/multi-label-text-classification-with-scikit-learn-30714b7819c5
Two well-known papers for this task are: http://nmis.isti.cnr.it/sebastiani/Publications/ACMCS02.pdf and http://www.jmlr.org/papers/volume2/lodhi02a/lodhi02a.pdf
- Badges
- Science topic
- Similar topics
- Data Mining
- Classification