مقایسهٔ روش‌ها

	شبکه عصبی کانولوشنی چندزبانه	شبکه LSTM چندزبانه
حوزه	یادگیری عمیق	یادگیری عمیق
خانواده	Machine learning	Machine learning
سال پیدایش	2014–2016	1997 (LSTM); multilingual NLP applications from ~2016
پدیدآور	Kim, Y. (seminal NLP CNN); multilingual extension by community	Hochreiter, S. & Schmidhuber, J. (LSTM base); multilingual application by the NLP community from ~2016
نوع	Deep learning classifier	Recurrent neural network (sequence model)
منبع بنیادین	Kim, Y. (2014). Convolutional Neural Networks for Sentence Classification. Proceedings of EMNLP 2014, pp. 1746–1751.	Hochreiter, S. & Schmidhuber, J. (1997). Long Short-Term Memory. Neural Computation, 9(8), 1735–1780.
نام‌های دیگر	ML-CNN, cross-lingual CNN, multilingual text CNN, multilingual ConvNet	Multilingual LSTM, Cross-lingual LSTM, Multi-language LSTM, Multilingual Recurrent Network
مرتبط	4	5
خلاصه	A Multilingual CNN applies convolutional filters over token embeddings drawn from two or more languages, producing shared feature representations that enable a single model to classify, tag, or extract information across language boundaries without training separate models per language. It extends the standard text-CNN architecture with multilingual or cross-lingual input embeddings.	A Multilingual LSTM is a Long Short-Term Memory recurrent network trained or fine-tuned to process sequences in multiple languages, typically by sharing a single model across language-specific or joint subword embeddings. It captures long-range dependencies in text and is applied to multilingual classification, named entity recognition, sentiment analysis, and sequence labeling.