Advanced Deep Learning Techniques for Real-Time Language Translation

  • Authors

    • Rohan Mirchandani Co-founder, Epigamia, India. Author

    DOI:

    https://doi.org/10.67228/30715628/IJMIET-2018PIIQ8L2

    Published 10-05-2018

  • Neural Machine Translation (NMT), Deep Learning, Real-Time Translation, Sequence-To-Sequence Models, Attention Mechanism, LSTM, RNN, Transformer, BLEU Score, Natural Language Processing

    Issue

    Vol. 1 No. 2 (2018)

    Section

    Articles

    How to Cite

    [1]
    R. Mirchandani, “Advanced Deep Learning Techniques for Real-Time Language Translation”, ijmiet, vol. 1, no. 2, pp. 01–13, Oct. 2018, doi: 10.67228/30715628/IJMIET-2018PIIQ8L2.

  • Abstract

    The unparalleled spreading of mobile technologies has strongly changed the way people interact with information and essentially changed the way people read without considering their demographics and cultures. Smart phone, tablet, and e-readers have transformed access, convenience, and customization of reading experience. This paper examines the implication of mobile technologies on reading habits in terms of cognitive engagement, frequency of reading, preference in reading content, and patterns of textual comprehension. Although mobile platforms have increased access to information democratically, there are other challenges that have been created by it, including fragmented attention, less deep reading, and dependence on short-form content. The study is done as a mixed-method study that includes survey, behavioral analysis and comparison of traditional and mobile reading set-ups. Results indicate the trend in the consumption pattern towards the nonlinear and skimming types of consumption. There is a growing preference of multimedia enhanced content that is interactive, thus affecting the understanding and retention. Moreover, mobile reading has raised the overall extent of reading but lowered the enduring concentration intervals. It also examines the generational variations and defines how younger apprehends adaptive digital reading habits whereas the older users tend to use hybrid reading. The place of algorithms and customized content delivery systems is also discussed where it can be seen how the recommendation systems shape reading decisions and are exposed to diversity. The paper ends by giving implications to the educators, publishers and technology developers and making the case that balanced reading system needs to be developed, which is why the advantages of mobile technologies should be integrated with the deep reading that should not be lost. The results are relevant to the current discussion on digital literacy and cognitive transformation in the mobile age.

  • References

    [1] I. Sutskever, O. Vinyals, and Q. V. Le, “Sequence to Sequence Learning with Neural Networks,” Advances in Neural Information Processing Systems (NeurIPS), vol. 27, pp. 3104–3112, 2014.

    [2] D. Bahdanau, K. Cho, and Y. Bengio, “Neural Machine Translation by Jointly Learning to Align and Translate,” International Conference on Learning Representations (ICLR), 2015.

    [3] K. Cho et al., “Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation,” Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 1724–1734, 2014.

    [4] M. Schuster and K. K. Paliwal, “Bidirectional Recurrent Neural Networks,” IEEE Transactions on Signal Processing, vol. 45, no. 11, pp. 2673–2681, 1997.

    [5] [5] S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997.

    [6] K. Cho, B. van Merriënboer, D. Bahdanau, and Y. Bengio, “On the Properties of Neural Machine Translation: Encoder–Decoder Approaches,” Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation (SSST), 2014.

    [7] J. Chung, C. Gulcehre, K. Cho, and Y. Bengio, “Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling,” NeurIPS Workshop, 2014.

    [8] Y. Wu et al., “Google’s Neural Machine Translation System: Bridging the Gap between Human and Machine Translation,” arXiv preprint arXiv:1609.08144, 2016.

    [9] T. Luong, H. Pham, and C. D. Manning, “Effective Approaches to Attention-based Neural Machine Translation,” Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 1412–1421, 2015.

    [10] M. Tu et al., “Modeling Coverage for Neural Machine Translation,” Association for Computational Linguistics (ACL), pp. 76–85, 2016.

    [11] R. Sennrich, B. Haddow, and A. Birch, “Neural Machine Translation of Rare Words with Subword Units,” Association for Computational Linguistics (ACL), pp. 1715–1725, 2016.

    [12] A. Vaswani et al., “Attention Is All You Need,” Advances in Neural Information Processing Systems (NeurIPS), vol. 30, pp. 5998–6008, 2017.

    [13] A. Graves, “Generating Sequences with Recurrent Neural Networks,” arXiv preprint arXiv: 1308.0850, 2013.

    [14] Z. Yang et al., “Hierarchical Attention Networks for Document Classification,” North American Chapter of the Association for Computational Linguistics (NAACL), pp. 1480–1489, 2016.

    [15] J. Gehring et al., “Convolutional Sequence to Sequence Learning,” International Conference on Machine Learning (ICML), pp. 1243–1252, 2017.

  • Downloads

Similar Articles

61-70 of 70

You may also start an advanced similarity search for this article.