This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
|
||||||||
|
Paper Details
Paper Title
Survey on question answering mechanism in real world call center applications
Authors
  Hemant P
Abstract
Machine reading comprehension has created a tremendous change in today’s Call-Centre Bots and question answering system. In real-time call centre environment aid to agents is very important as it helps to solve complex problems and in turn improve customer satisfaction. Machine Reading Comprehension can help in understanding customer issues after running analysis on the speech to text output and provide right set of questions to be asked to the customer so that the conversation can be channeled properly. Due to this the average handling time would reduce by a substantial number. We have made a comparison of different system build on MRC and parameters like the dataset used for training and validation, performance on various benchmarks and features. So according to the research done it has been observed that the question answering system has been shifted from query matching to read the exact passage to generate the answers. The datasets which were used for training and how the performance differs according to the dataset and the training mechanism were also discussed. Then the evaluation and performance of all the tasks are compared and best algorithm was chosen. The use case on which the specific algorithm was working is also important. In this way the final algorithm was concluded by observing all these factors.
Keywords- Generative Chatbot, Machine Reading Comprehension, Question Answering
Publication Details
Unique Identification Number - IJEDR2003061Page Number(s) - 422-425Pubished in - Volume 8 | Issue 3 | August 2020DOI (Digital Object Identifier) -    Publisher - IJEDR (ISSN - 2321-9939)
Cite this Article
  Hemant P,   "Survey on question answering mechanism in real world call center applications", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.8, Issue 3, pp.422-425, August 2020, Available at :http://www.ijedr.org/papers/IJEDR2003061.pdf
Article Preview
|
|
||||||
|