Deep learning applications and challenges in big data analytics

Deep learning applications and challenges in big data analytics


EOI: 10.11242/viva-tech.01.05.MCA_18

Download Full Text here



Citation

Prof. Neha Lodhe, Mr. Sumit Bhatkar, Ms. Neha Tiwari, "Deep learning applications and challenges in big data analytics", VIVA-IJRI Volume 1, Issue 6, Article MCA_18, pp. 1-6, 2023. Published by MCA Department, VIVA Institute of Technology, Virar, India.

Abstract

Two areas of data science with a lot of interest are big data analytics and deep learning. Big Data has grown in importance as a result of the large-scale collection of domain-specific data by both public and private entities, which can provide useful information regarding issues like national intelligence, cyber security, fraud detection, marketing, and medical informatics. Large data sets are being analysed by businesses like Google and Microsoft for business analysis and decisions that will affect both current and future technologies. Via a hierarchical learning process, deep learning algorithms extract high-level, complex abstractions as data representations. Based on relatively simpler abstractions created in the previous level of the hierarchy, complex abstractions are learned at a given level. Massive amounts of unsupervised data can be analysed and learned from using deep learning, which makes big data analytics possible even when the raw data is largely unlabeled and uncategorized. In this work, we investigate how Deep Learning might be used to solve certain key issues in Big Data Analytics, such as extracting intricate patterns from enormous amounts of data, semantic indexing, data tagging, quick information retrieval, and simplification of discriminative tasks. We also look into several Deep Learning research areas that require more investigation in order to address specific Big Data Analytics difficulties, such as streaming data, high-dimensional data, model scalability, and distributed computing. Defining data sample criteria, domain adaption modelling, establishing criteria for generating meaningful data abstractions, enhancing semantic indexing, semi-supervised learning, and active learning are some of the problems we pose in our conclusion to provide insights into pertinent future studies.

Keywords

Big data, Data Analytics, Data Mining, Deep learning, Machine Learning

References

  1. Dominicos P (2012) A few important facts concerning machine learning. 55 Commun ACM (10)
  2. Bengio Y, LeCun Y (2007) Scaling learning algorithms towards, AI. In: Bottou L, Chapelle O, DeCoste D, Weston J (eds). Large Scale Kernel Machines. MIT Press, Cambridge, MA Vol. 34. pp 321–360. http://www.iro.umontreal.ca/~ lisa/pointeurs/bengio+lecun_chapter2007.pdf
  3. Rose DC, Arel I, and Karnowski TP (2010) A fresh field of study in artificial intelligence is deep machine learning. 5:13–18 IEEE Comput Intell
  4. What Is Big Data, by E. Dumbill (2012)? An introduction to the landscape of big data. In: Making Data Work, Strata 2012. Santa Clara, California: O'Reilly O’Reilly
  5. TM Khoshgoftaar (2013) overcoming the hurdles of big data. In: 25th International Conference on Software Engineering and Knowledge Engineering Proceedings, Boston, Massachusetts. ICSE. Keynote Speaker invited
  6. Y. Bengio, Learning Deep Architectures for AI, 2009. Hanover, Massachusetts-based Now Publishers Inc.
  7. Bengio Y. (2013). Deep learning of representations: A look ahead. First International Conference on Statistical Language and Speech Processing Proceedings. Springer, Tarragona, Spain. SLSP'13. Pages 1–37. http://dx.doi.org/10.1007/978-3-642-39593-2 1
  8. What Is Big Data, asks Dumbill E. (2012)? An introduction to the landscape of big data. In: Making Data Work, Strata 2012. Santa Clara, California: O'Reilly O’Reilly
  9. Grobelnik M., [9] (2013) Tutorial for Big Data. Forum for European Data. http://www.slideshare.net/EUDataForum/edf2013-bigdatatutorialmarkogrobelnik?related=1
  10. Scaling learning algorithms towards AI. Bengio and LeCun. In: Weston J, DeCoste D, Chapelle O, Bottou L (eds). Kernels on a large scale. Cambridge, MA: MIT Press Vol. 34. pages 321-360. lisa/pointeurs/bengio+lecun chapter2007.pdf at www.iro.umontreal.ca
  11. Dahl, GE, Hinton, and Mohamed A-R (2012) deep belief network-based acoustic modelling. Procedure for Audio Speech Lang. IEEE Trans, 20(1), pp. 14–22
  12. Zhou G, Sohn K, Lee H (2012) Online incremental feature learning with denoising autoencoders. In: International Conference on Artificial Intelligence and Statistics. JMLR.org. pp 1453–1461
  13. Weinberger KQ, Chen M, Xu ZE, Sha F (2012) Autoencoders with marginalised denoising for domain adaptation. In: Edingburgh, Scotland, 29th International Conference on Machine Learning Proceedings
  14. Coates A, Ng A (2011) The significance of vector quantization and sparse coding for encoding as opposed to training. In: The 28th International Conference on Machine Learning Proceedings. p. 921–928 in Omnipr
  15. Domain adaptation for large-scale sentiment classification: A deep learning approach, Glorot X, Bordes A, Bengio Y (2011). pp. 513-520 58 in: Proceedings of the 28th International Conference on Machine Learning.
  16. Chopra S, Balakrishnan S, Gopalan R (2013) Dlid: Deep learning for domain adaptation by interpolating between domains. In: Workshop on Challenges in Representation Learning, Proceedings of the 30th International Conference on Machine Learning, Atlanta, GA