השוואת שיטות

	ניתוח סנטימנט סמי-מפוקח	ניתוח סנטימנט בלמידה עצמית-פיקוח
תחום	למידה עמוקה	למידה עמוקה
משפחה	Machine learning	Machine learning
שנת המקור	2002–2008	2019–present
הוגה השיטה	Zhu, X.; Pang, B. & Lee, L. (foundational works)	Devlin et al. (BERT paradigm); extended by Sun et al. and others
סוג	Semi-supervised classification	Pre-train then fine-tune NLP pipeline
מקור מכונן	Zhu, X. (2005). Semi-Supervised Learning Literature Survey. Technical Report 1530, Computer Sciences, University of Wisconsin-Madison.	Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019). BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of NAACL-HLT 2019 (pp. 4171–4186). Association for Computational Linguistics.
כינויים	SSSA, semi-supervised opinion mining, label-propagation sentiment classification, self-training sentiment analysis	SSL-based sentiment analysis, self-supervised opinion mining, pre-training for sentiment, unsupervised pre-training sentiment
קשורות	4	2
תקציר	Semi-supervised sentiment analysis combines a small set of manually labeled text samples with a large pool of unlabeled text to train opinion classifiers. By propagating sentiment signals from labeled seeds to unlabeled data through self-training, label propagation, or consistency regularization, the approach achieves competitive accuracy without the cost of labeling large corpora.	Self-supervised sentiment analysis combines large-scale unsupervised pre-training — through objectives such as masked language modeling or contrastive prediction — with fine-tuning on a small labeled sentiment corpus. The approach, popularized by BERT and its variants, dramatically reduces the need for hand-labeled data while achieving state-of-the-art accuracy on positive/negative/neutral opinion classification tasks.