Journal
ELECTRONICS
Volume 10, Issue 12, Pages -Publisher
MDPI
DOI: 10.3390/electronics10121412
Keywords
BERT; word and sentence transformers; monolingual and cross-lingual experiments; EN; DE; FR; LT; LV; PT languages
Categories
Funding
- European Regional Development Fund within the joint project of SIA TILDE
- University of Latvia Multilingual Artificial Intelligence Based Human Computer Interaction [1.1.1.1/18/A/148]
- European Union [825081]
Ask authors/readers for more resources
This research experimentally solved the intent detection problem in five target languages using an English dataset. By utilizing various models and methods, they overcame the data scarcity issue and demonstrated the robustness of sentence transformers under different cross-lingual conditions.
Due to recent DNN advancements, many NLP problems can be effectively solved using transformer-based models and supervised data. Unfortunately, such data is not available in some languages. This research is based on assumptions that (1) training data can be obtained by the machine translating it from another language; (2) there are cross-lingual solutions that work without the training data in the target language. Consequently, in this research, we use the English dataset and solve the intent detection problem for five target languages (German, French, Lithuanian, Latvian, and Portuguese). When seeking the most accurate solutions, we investigate BERT-based word and sentence transformers together with eager learning classifiers (CNN, BERT fine-tuning, FFNN) and lazy learning approach (Cosine similarity as the memory-based method). We offer and evaluate several strategies to overcome the data scarcity problem with machine translation, cross-lingual models, and a combination of the previous two. The experimental investigation revealed the robustness of sentence transformers under various cross-lingual conditions. The accuracy equal to similar to 0.842 is achieved with the English dataset with completely monolingual models is considered our top-line. However, cross-lingual approaches demonstrate similar accuracy levels reaching similar to 0.831, similar to 0.829, similar to 0.853, similar to 0.831, and similar to 0.813 on German, French, Lithuanian, Latvian, and Portuguese languages.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available