CFP last date
22 July 2024
Call for Paper
August Edition
IJCA solicits high quality original research papers for the upcoming August edition of the journal. The last date of research paper submission is 22 July 2024

Submit your paper
Know more
The week's pick
Responsive image
A semi-parallel data acquisition method in electrical impedance tomography using undersampling technique

Adriano Regis Daniel J. Pagano Francisco R.M. Mota Marduck M. Henao
Random Articles
Reseach Article

Hybrid Machine Learning Approach for Task-Oriented Dialog Systems

by Ganesh Reddy Gunnam, Devasena Inupakutika, Rahul Mundlamuri, Sahak Kaghyan, David Akopian
journal cover thumbnail

International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
Volume 186 - Number 23
Year of Publication: 2024
Authors: Ganesh Reddy Gunnam, Devasena Inupakutika, Rahul Mundlamuri, Sahak Kaghyan, David Akopian
10.5120/ijca2024923679

Ganesh Reddy Gunnam, Devasena Inupakutika, Rahul Mundlamuri, Sahak Kaghyan, David Akopian . Hybrid Machine Learning Approach for Task-Oriented Dialog Systems. International Journal of Computer Applications. 186, 23 ( May 2024), 35-42. DOI=10.5120/ijca2024923679

@article{ 10.5120/ijca2024923679,
author = { Ganesh Reddy Gunnam, Devasena Inupakutika, Rahul Mundlamuri, Sahak Kaghyan, David Akopian },
title = { Hybrid Machine Learning Approach for Task-Oriented Dialog Systems },
journal = { International Journal of Computer Applications },
issue_date = { May 2024 },
volume = { 186 },
number = { 23 },
month = { May },
year = { 2024 },
issn = { 0975-8887 },
pages = { 35-42 },
numpages = {9},
url = { https://ijcaonline.org/archives/volume186/number23/hybrid-machine-learning-approach-for-task-oriented-dialog-systems/ },
doi = { 10.5120/ijca2024923679 },
publisher = {Foundation of Computer Science (FCS), NY, USA},
address = {New York, USA}
}
%0 Journal Article
%1 2024-05-31T22:32:03.063825+05:30
%A Ganesh Reddy Gunnam
%A Devasena Inupakutika
%A Rahul Mundlamuri
%A Sahak Kaghyan
%A David Akopian
%T Hybrid Machine Learning Approach for Task-Oriented Dialog Systems
%J International Journal of Computer Applications
%@ 0975-8887
%V 186
%N 23
%P 35-42
%D 2024
%I Foundation of Computer Science (FCS), NY, USA
Abstract

Nowadays, automated chatbots are commonly used since they easily provide essential information. While generic chatbots are essential for open-domain dialog, specific applications are better served with task-oriented dialog systems. These task-oriented dialog systems typically solve particular tasks in the application where the chatbot and user know what they are discussing (both sides know the scope and context of the conversation topic). The majority of these chatbots work based on keywords. Keyword extraction has been a well-established field in the natural language processing (NLP) domain for quite some time. It is crucial in various applications, such as information retrieval, search engine optimization, and content summarization. Recently, there has been a growing interest in the contextual recognition of keywords, which aims to identify keywords in a given text based on their contextual relevance. Additionally, integrating Large Language Models (LLMs) with intent prediction (IP) has opened new possibilities for interpreting and utilizing keywords in a more context-aware manner. In particular, one such LLM, BERT, a SQuAD dataset-based NLP model, has become a popular question-answer set. However, task-oriented systems still challenge specific questions, such as yes/no and synonym-based inquiries. Thus, a hybrid model involving LLMs and IP merits additional study. This paper explores the intersection of keyword extraction, LLMs, and Intent Prediction in the context of protocol-driven chatbots, particularly those designed for task-oriented applications, emphasizing their potential in addressing a niche application. Specifically, this paper presents a hybrid approach (TaskBERT) that addresses these challenges. The evaluation results demonstrate that TaskBERT outperforms Google Dialogflow and the performant keyword extraction tool KeyBERT.

References
  1. M. Nakano and K. Komatani, “A framework for building closed-domain chat dialogue systems,” Knowledge Based Systems, vol. 204, p. 106212, Sep. 2020, doi: 10.1016/j.knosys.2020.106212.
  2. P. H. Saurav, A. R. Limon, R. Amin, and M. S. Rahman, “Multi-Layer Open-Domain Bangla Conversational Chatbot with a Hybrid approach,” 2023 International Conference on Next-Generation Computing, IoT and Machine Learning (NCIM), Jun. 2023, doi: 10.1109/ncim59001.2023.10212816.
  3. F. Cui, Q. Cui, and Y. Song, “A Survey on Learning-Based Approaches for Modeling and Classification of Human–Machine Dialog Systems,” IEEE Transactions on Neural Networks and Learning Systems, vol. 32, no. 4, pp. 1418–1432, Apr. 2021, doi: 10.1109/tnnls.2020.2985588.
  4. S. Rose, D. Engel, N. Cramer, and W. Cowley, “Automatic Keyword Extraction from Individual Documents,” in Text mining: applications and theory, Wiley online library, 2010. doi: 10.1002/9780470689646.ch1.
  5. S. Qaiser and R. Ali, “Text mining: Use of TF-IDF to examine the relevance of words to documents,” International Journal of Computer Applications, vol. 181, no. 1, pp. 25–29, Jul. 2018, doi: 10.5120/ijca2018917395.
  6. S. Pan, Z. Liu, and J. Dai, “An improved TextRank keywords extraction algorithm,” Proceedings of the ACM Turing Celebration Conference - China, May 2019, doi: 10.1145/3321408.3326659.
  7. A. Andy, M. Robert, and M. F. Chouikha, “Exploiting synonyms to improve question and answering systems,” International Journal of Computer Applications, vol. 108, no. 18, pp. 24–27, Dec. 2014, doi: 10.5120/19012-0523.
  8. M. Grice and M. Savino, “Information structure and questions: evidence from task-oriented dialogues in a variety of Italian,” in Regional variation in intonation, P. Gilles and J. Peters, Eds. 2004. [Online]. Available: https://www.cs.columbia.edu/~julia/papers/grice&savino04.pdf
  9. G. N, V. G, and T. A. Vinetia, “Intent Classification using BERT for Chatbot application pertaining to Customer Oriented Services,” International Conference on Combinatorial and Optimization, ICCAP, Dec. 2021, doi: 10.4108/eai.7-12-2021.2314563.
  10. N. Sabharwal and A. Agrawal, “Introduction to Google Dialogflow,” in Apress eBooks, 2020, pp. 13–54. doi: 10.1007/978-1-4842-5741-8_2.
  11. “Microsoft Luis,” Language Understanding (LUIS). https://www.luis.ai/ (accessed Dec. 01, 2023).
  12. “Amazon,” Amazon Lex. https://aws.amazon.com/lex (accessed Dec. 01, 2023).
  13. F. You, S. Zhao, and J. Chen, “A topic information fusion and semantic relevance for text summarization,” IEEE Access, vol. 8, pp. 178946–178953, Jan. 2020, doi: 10.1109/access.2020.2999665.
  14. N. Firoozeh, A. Nazarenko, F. Alizon, and B. Daille, “Keyword extraction: Issues and methods,” Natural Language Engineering, vol. 26, no. 3, pp. 259–291, Nov. 2019, doi: 10.1017/s1351324919000457.
  15. S. Beliga, A. Meštrović, and S. Martinčić-Ipšić, “An Overview of Graph-Based Keyword Extraction Methods and Approaches,” DOAJ (DOAJ: Directory of Open Access Journals), Jul. 2015, [Online]. Available: https://doaj.org/article/c60517233bf44eae8807eaba0a2ebf59
  16. M. Grootendorst, “KeyBERT: Minimal keyword extraction with BERT,” Zenodo, 2010, doi: 10.5281/zenodo.4461265.
  17. I. Alberts et al., “Large language models (LLM) and ChatGPT: what will the impact on nuclear medicine be?,” European Journal of Nuclear Medicine and Molecular Imaging, vol. 50, no. 6, pp. 1549–1552, Mar. 2023, doi: 10.1007/s00259-023-06172-w.
  18. C. Raffel et al., “Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer,” Journal of Machine Learning Research, vol. 21, no. 140, pp. 1–67, Jan. 2020, [Online]. Available: https://jmlr.org/papers/volume21/20-074/20-074.pdf
  19. A. Bhargava, A. Çelikyılmaz, D. Hakkani‐Tür, and R. Sarikaya, “Easy contextual intent prediction and slot detection,” IEEE International Conference on Acoustics, Speech and Signal Processing, May 2013, doi: 10.1109/icassp.2013.6639291.
  20. Q. Chen, Z. Zhu, and W. Wang, “BERT for joint intent classification and slot filling,” arXiv (Cornell University), Feb. 2019, [Online]. Available: https://arxiv.org/pdf/1902.10909.pdf
  21. M. Huggins, S. Alghowinem, S. Jeong, P. Colón-Hernández, C. Breazeal, and H. W. Park, “Practical Guidelines for Intent Recognition,” ACM/IEEE International Conference on Human-Robot Interaction, Mar. 2021, doi: 10.1145/3434073.3444671.
  22. P. Rajpurkar, J. Zhang, K. Lopyrev, and P. Liang, “SQuAD: 100,000+ Questions for Machine Comprehension of Text,” arXiv (Cornell University), Jun. 2016, doi: 10.48550/arxiv.1606.05250.
  23. Y. Xiao, “A Transformer-based Attention Flow Model for Intelligent Question and Answering Chatbot,” International Conference on Computer Research and Development (ICCRD), Jan. 2022, doi: 10.1109/iccrd54409.2022.9730454.
  24. “Thesaurus.com - The world’s favorite online thesaurus!,” Thesaurus.com, Dec. 01, 2023. https://www.thesaurus.com/ (accessed Dec. 01, 2023).
  25. “Synonym Finder,” WordHippo. https://synonym.wordhippo.com/ (accessed Dec. 01, 2023).
  26. “ChatGPT.” https://chat.openai.com (accessed Dec. 01, 2023).
  27. W. Cai, Y. Jin, and L. Chen, “Task-Oriented user evaluation on Critiquing-Based recommendation chatbots,” IEEE Transactions on Human-Machine Systems, vol. 52, no. 3, pp. 354–366, Jun. 2022, doi: 10.1109/thms.2021.3131674.
  28. “squad · Datasets at Hugging Face,” Apr. 06, 2001. https://huggingface.co/datasets/squad (accessed Dec. 01, 2023).
  29. M. Q. Khan et al., “Impact analysis of keyword extraction using contextual word embedding,” PeerJ, vol. 8, p. e967, May 2022, doi: 10.7717/peerj-cs.967.
  30. “Pretrained Models Sentence Transformers documentation.”https://www.sbert.net/docs/pretrained_models.html (accessed Dec. 01, 2023).
Index Terms

Computer Science
Information Sciences

Keywords

artificial intelligence natural language processing closed domain chatbot intent prediction

Powered by PhDFocusTM