NLP Models Behind RASA Stack
Ajith Shenoy1, Sushma Ravindra Y2, Akash Sharma3, Akshay Rajan4, Akshay GV5
1Ajith Shenoy, School of C&IT, REVA University, India.
2Sushma Ravindra Y, School of C&IT, REVA University, India.
3Akash Sharma, School of C&IT, REVA University, India.
4Akshay Rajan, School of C&IT, REVA University, India.
5Akshay GV, School of C&IT, REVA University, India.
Manuscript received on 04 June 2019 | Revised Manuscript received on 12 June 2019 | Manuscript Published on 29 June 2019 | PP: 62-66 | Volume-8 Issue-5S, May 2019 | Retrieval Number: E10130585S19/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This paper brings about the foundation of a platform for conversational AI the Rasa platform. This Rasa stack contains a block of open source machine learning tools exclusively used in intend to create a contextual chatbots and assistants. The services hold by this platform undergoes a major classification of powerful APIs and embedded together with Rasa stack which includes Rasa core and Rasa NLU in the form of an event stream discussed throughout this paper and also the algorithm involved in building upon this platform. Its ingredients include the Bag of words algorithm helping in simplifying representation used in the NLP, CRFs – Conditional Random Field used in statistical modelling and machine learning platforms and also advanced technology such as LTSM neural networks. This paper discusses all the algorithms involved in building up the platform and also the result produced in building up the student assistant chatbot using this platform. It also encourages the use of this RASA platform for the user required custom format as per their requirements and also promotes to contribute in developing the platform for better efficiency of the platform to function.
Keywords: Bag of words, Chatbot, CRFs, NLP, Rasa Stack.
Scope of the Article: Probabilistic Models and Methods