Traditional Machine Learning and Deep Learning-based Text Classification for Turkish Law Documents using Transformers and Domain Adaptation.
Onur AkçaGiyaseddin BayrakAbdul Majeed IssifuMurat Can GanzPublished in: INISTA (2022)
Keyphrases
- text classification
- deep learning
- domain adaptation
- document classification
- machine learning
- labeled data
- text documents
- sentiment classification
- unsupervised learning
- text categorization
- unlabeled data
- co training
- transfer learning
- text mining
- semi supervised learning
- bag of words
- machine learning algorithms
- sentiment analysis
- feature selection
- cross domain
- n gram
- target domain
- information extraction
- learning algorithm
- decision trees
- multi label
- information retrieval
- multiple sources
- support vector machine
- knowledge representation
- document clustering
- data analysis
- active learning
- general purpose
- similarity measure
- reinforcement learning
- cross lingual
- xml documents
- data sources
- semantic features