Crowdsourced NLP Retraining Engine in Chatbots


Chatbots became a common customer service and e-commerce tool in recent years. Most of the chat platforms offer chatbot integration opportunities. Natural Language Processing (NLP) are powering chatbots worldwide and are at a well-developed stage for English and major languages. Still, lots of improvement and regular retraining required as use cases for chatbots can be very different and evolving every day. Retraining is also important for languages that don’t have very strong NLP. In this paper, a crowdsourced, user-generated retraining system is being suggested to make the chatbot NLPs auto-retrained by the users who are actually using it. Experiments result in an average of 17% improvement in the NLP results.

About this paper

Cite this paper

Nezami, M.A., Rukham, R. (2022). Crowdsourced NLP Retraining Engine in Chatbots. In: Al-Emran, M., Al-Sharafi, M.A., Al-Kabi, M.N., Shaalan, K. (eds) Proceedings of International Conference on Emerging Technologies and Intelligent Systems. ICETIS 2021. Lecture Notes in Networks and Systems, vol 322. Springer, Cham.

Download citation