Authors: Dr. G. Niranjana , DJNV Sai Trinadh
DOI Link: https://doi.org/10.22214/ijraset.2022.40014
Certificate: View Certificate
A chat bot is an AI tool that interprets information provided by humans and responds accordingly. It is used to carry out tasks like mimic human behavior. Many of the chat bot conversations between humans and machines are repetitive. Instead of just talking to one another, a chat bot can communicate with the outside world by analyzing the user\'s needs and actions. Having a chat bot assistant that can help users complete their tasks makes them feel like they\'re doing something right. It eliminates the user\'s need to do the work for them. This idea has been proposed to introduce a dual-agent system that will allow users to interact with the chat bot without leaving the human interaction. It can make decisions on behalf of the user, which reduces the user\'s efforts in carrying out a task. This concept is demonstrated in our study. The concept is to create a chat bot that can diagnose a health issue & provide basic information before reaching out to a doctor.
A chat bot is an artificial intelligence system that works seamlessly by acting like a human. It helps the user complete their task by delivering the most accurate and up-to- date information. Each chat bot has its own unique characteristics, which it uses to respond to the needs of users. It tries to fulfil the expectations of the users.
A chat bot is a type of AI system that helps users to perform various tasks without requiring them to interact with a human. It uses algorithms to collect and interpret the data it receives.
Most of the time, a chat bot and a user are having a repetitive conversation. In most cases, the user has to enter all the usual data that the user would typically enter in a regular conversation.
The proposed system helps users save time and effort by communicating with another chat bot on their behalf. It learns about the user’s needs and wants and takes appropriate decisions on their behalf. The rise of digital healthcare has brought about a paradigm shift in the way patients manage their conditions. This transformation aims at providing personalized Healthcare services and making them self-aware. The main benefits of using a conversation Healthcare chat bot are ease of use and accessibility.
II. RELATED WORK
Our system is powered by the amazing work of RASA. Through its NLU, which is a natural language, it can easily understand and convert complex sentences into logical representations. Each intent has a parameter that denotes the probability that the given goal will be met. This parameter is called confidence.
Reference  Both NLU and NLG use a similar technique called collaborative reinforcement technique to help minimize the load on the user by remembering the past conversations. This method is very useful for minimizing the user’s load.
This chat bot has a single agent that can ask any question to it. It uses NLP to analyse the message sent by the user and determine its intent. It will use Natural language processing to identify its intent by analysing the message.
Chat bots are an integral part of our day to day lives and how they can substitute humans in various fields. With the evolution of chat bots, we are now able to provide a more complete and efficient service to our customers. They have closely studied how chat bots have improved and analysed what could be the future of chat bots. With the evolution of chat bots, we can now create a truly conversational agent that will allow people to do more tasks without requiring much effort on their part .
Reference  This paper presents an recognition approach that uses multi-modal approach. They have collected corpus of terms and expressions to learn and represent them using word vector. These studies are based on various AI methods to classify emotions. The studies train emotion classifications models using various AI techniques. It aims to develop a multi-modal approach to understand users' dialogues using NLP and NLG. The concept of emotion-recognition is used to collect corpuses of words and determine their semantic importance the concept of simultaneous learning was discussed to explain how two agents can communicate with each another without a simulated user.
How they improve concurrently despite having a simulated user which minimizes the agent’s interaction with the user will improve during communication among them was discussed in the paper .
A web-based chat bot is developed, which is equipped with a voice-based recognition system. The chat bot is created as web-friend based on text. This project involves analysing and capturing an input signal to trigger a voice recognition system. The server used in this project is a black box approach based on SOAP .
This chat bot aims to communicate with humans. It stores the knowledge about the given sentence and comes up with a decision to answer the question. The RDBMS is used to store the knowledge of the chat bot. This chatbot learns about a given sentence and comes up with an answer based on its knowledge. The chatbot will also score the similarity of the given sentence. .
The chat bot implemented a pattern comparison method to save the order of the sentence in the input. It sends a response to the database after choosing a response from the list of available options. They have taken the input using text function and other punctuation .
A chat bot created using the same programming language as an framework that provides chat bot services to the users. It has been developed with a web-based chat bot API. The platform provides a wide range of features that allow users to simulate real- world problem-solving tasks. It also supports interactive problem-solving sessions .
III. PROBLEM DEFINATION
There has been a growing interest in the use of chatbot technology for the provision of healthcare services. This paper aims to provide a deeper understanding of how these emerging platforms address the various aspects of the healthcare service provision. Chat bots in healthcare could help patients manage their medical conditions and improve their access to care. They could also provide them with immediate medical information. Chat bots are tools that can help users with healthcare information such as disease diagnosis and treatment. This will have a deep impact on the idea to reduce healthcare costs and provide better accessibility to medical knowledge through healthcare chatbot. With help of RASA-NLU we tried come up with the model for the addressed concerns.
IV. PROPOSED SYSTEM
The chatbot is developed for a specific domain where the user has not interacted with the system before. Once the system learns about the user's requirements, the chatbot takes the necessary actions. The user can either make a new interaction or the system can interact on his/her behalf. Socket Connection is a connection b/w the server and the client. It allows the server to communicate with the client without having to track it.
The system is partitioned into three parts:
1. Ionic or flask (tentative), 2. Server, and 3.MongoDB.
The user goes through the steps mentioned to get started with the system. After that, the user can start working with the interface by providing basic details. The user can interact with the system by entering all the questions and queries related to the health. The system then interprets the query or questions as a series of words that the user wants.
The server that is listening and emitting events for the chat needs a connection to the network. This is done through a npm module that is called npm-socket-io. On the Security front, for secure communication between client and the server, HTTPis used.
B. Server and Conversational Agent
The main function of any chatbot is to take the user's requirements and provide them with the most relevant information. This part of the system works seamlessly if the user provides the required specifications for a new interaction.
When the conversation is completed successfully, the extraction of entities takes place and then gets stored in the database. In case the user wants to speak to the system on user behalf, the system takes help of the previous interactions to make the interaction more effective.
The server & the second agent carry out the conversation to get the value that the user has requested. The resulting value is then returned to the user.
The NLU model is an open-source version of the NLP framework that provides a complete set of entities and intents that can be trained using predefined terms. This model is also more efficient to use for NLU. The server is the main component that operates the system. Its goal is to understand the message sent by the user and react in according with the help of RASA. It uses reinforcement learning and probabilistic models to learn how to respond efficiently to the user's input. The model is to be thoroughly trained to understand the language. The server then tries to answer the second agent's questions with defined intents. The model then generates the text that the second agent needs to understand. NLG is a technique used to generate responses by the server based on the intent of the user. The verb part of the response will be derived from the preference order's noun part. This method generates the noun and vice versa parts of the response. A server job comes with periodically monitoring the database to update the values of all the interactions entities that are associated with them. This ensures that the user can interact with the system favoured with the preference of the user. The preference analysis for each interaction is performed to give the best results. This method helps the user interaction more efficiently. The RESTful APIs manage the sever-side functions. The chatbot's events are managed through the same server. The Python-shell is used to run the pipeline processes for the chatbot's functionalities. Socket events are used to control the fluency of the interaction b/w the two agents in the event of preference order.
C. MongoDB (Database)
Mongo DB is a high-performance, object- oriented database platform that allows developers to create and consume data without having to manage complex relationships with the data. The users and their data will be stored in one database, the information requested will be stored in the other. Its unique feature is that it gives the users complete freedom from the constraints of SQL. For instance, if a user has interacted regarding COVID-19 multiple times, then the preference of the user will be COVID- 19 which is more frequently used.
V. FUTURE SCOPE
The chatbot can easily perform various tasks with the help of predefined rules and procedures. It can also learn more about the user and make more informed decisions based on its findings. The next step for this project can be a emotional tracker which can be used to help not just users with health issues (physical illness) but also those who are suffering from mental stability, I believe chatbots can be a revolutionary idea if trained well they can be a psychologic solution for people who are introverts which helps them that the fact that they are talking or chatting with a bot.
Despite the advantages of chatbot technology, healthcare organizations are still not able to capitalize on the opportunities it can provide completely. We can aim to identify the various roles of chatbots and to provide a framework for designing and implementing effective chatbot services for the concerns mentioned above.
We proposed a system in this paper which can adapt to the user characteristics and take the entities of their previous interactions as a training data in a sense, while a new user interacts with our system, It will be totally in users hands to communicate for the first few interactions, later as the interactions go on the preferences from the user previous data will be collected and used to minimize the user need to communicate and provide with best appropriate responses. As we are focusing on healthcare department, if a user who had already interacted with the system previously will have the facility to choose to interact with the system completely by themselves or the user can opt for 2nd conversational agent to do the interaction on his/her behalf. For example, a user who already interacted for 5 times of which he/she used four interactions to query about symptoms of corona virus, that would become his/her preference and them when they interact again, he/she can just choose to be answered on the previous query basis. Which will eliminate the whole process of explanation happened in the previous interactions. We took help of RASA-NLU for this process in which after the user enters their message , intents and entities will be extracted , which will get further processed by dialogue management where tracker and slots sections take place later in the next step the message generator comes into play after which by users choice2nd conversational agent takes the message generated and tries to simply it more using the entities extracted from the previous interactions of the user, then finally the message will be displayed for the user on the display.
 Tom Bocklisch, Joey Faulkner, Nick Pawlowski, Alan Nichol, “RASA: Open Source Language Understanding and Dialogue Management” 2019.  Alexandros Papangelis, Yi-Chia Wang, Piero Molino, Gokhan Tur”, Collaborative Multi-Agent Dialogue Model Training Via Reinforcement Learning” Uber AI San Francisco, California.  Aafiya Shaikh1, Dipti More2, Ruchika Puttoo3, Sayli Shrivastav4, Swati Shinde4, ”A Survey Paper on Chatbots ”International Research Journal of Engineering and Technology (IRJET) Apr 2019.  K. Oh, D. Lee, B. Ko and H. Choi, \"A Chatbot for Psychiatric Counselling in Mental Healthcare Service Based on Emotional Dialogue Analysis and Sentence Generation,\" 2018 18th IEEE International Conference on Mobile Data Management (MDM), Daejeon, 2018, pp. 371-375. doi: 10.1109/MDM.2017.64  Neelkumar P. Patel, Devangi R. Parikh, “AI and Web-Based Human-Like Interactive University Chat bot (UNIBOT)”, IEEE,2019.  Nitirajsingh Sandu , Ergun Gide, “Adoption of AI- Chat bots to Enhance Student Learning Experience in Higher Education in India”, IEEE ,2019.  H. N. Io, C. B. Lee, \"Chatbots and conversational agents: A bibliometric analysis\" 2019 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM).  Urmil Bharti, Deepali Bajaj, Hunar Batra, Shreya Lalit, Shweta Lalit, Aayushi Gangwan, “Med bot: Conversational Artificial Intelligence Powered Chat bot for Delivering Tele-Health after COVID-19”,IEEE , 2020  Urmil Bharti, Deepali Bajaj, Hunar Batra, Shreya Lalit, Shweta Lalit, Aayushi Gangwan, “Med bot: Conversational Artificial Intelligence Powered Chat bot for Delivering Tele-Health after COVID-19”,IEEE , 2020.  Prakhar Srivastava, Automatized Medical Chat IEEE, 2020.  Parth Thosani, Manas Nishant Singh,” bot (Medibot)”, Sinkar, Jaydeep Vaghasiya, Radha Shankarmani,” A Self Learning Chat-Bot from User Interactions and Preferences”, IEEE , 2020.  Vladimir Ilievski1, Claudiu Musat2, Andreea Hossmann2, Michael Baeriswyl2 “Goal-Oriented Chatbot Dialog Management Bootstrapping with Transfer Learning ”1 School of Computer and Communication Sciences, EPFL, Switzerland 2 Artificial Intelligence Group - Swisscom AG 2018  Diksha Khurana1, Aditya Koli1, Kiran Khattar 1,2 and Sukhdev Singh1“Natural Language Processing: State of The Art, Current Trends and Challenges” Manav Rachna International University2018.  John Bateman, Michael Zock ”Natural Language Generation(Cognitive, linguistic, social dimensions)”, French National Centre for Scientific Research 2019.  Nurgalieva, L., Baez, M., Adamo, G., Casati, F., and Marchese, M., \"Designing interactive systems to mediate communication between formal and informal caregivers in aged care\" in IEEE Access, 7, 171173- 171194, 2019.  Montenegro, J. L. Z., da Costa, C. A., and da Rosa Righi, R., \"Survey of conversational agents in health\" in Expert Systems with Applications, vol. 129, pp. 56–67, 2019.  Fitzpatrick, K. K., Darcy, A., and Vierhile, M., \"Delivering cognitive behavior therapy to young adults with symptoms of depression and anxiety using a fully automated conversational agent (Woebot): a randomized controlled trial\" in Journal of Medical Internet Research Mental Health, vol. 4, no. 2, 2017.  Hwang, I., Lee, Y., Yoo, C., Min, C., Yim, D., and Kim, J., \"Towards Interpersonal Assistants: Next-Generation Conversational Agents\" in IEEE Pervasive Computing, vol. 18, no. 2, pp. 21-31, 2019.  Fadhil, A., and Schiavo, G., \"Designing for health chatbots\" arXiv:1902.09022, 2019.  Harms, J. G., Kucherbaev, P., Bozzon, A., and Houben, G. J., \"Approaches for Dialog Management in Conversational Agents\" in IEEE Internet Computing, vol 23, no. 2, pp. 13-22, 2019.  Nurgalieva, L., Baez, M., Adamo, G., Casati, F., and Marchese, M., \"Designing interactive systems to mediate communication between formal and informal caregivers in aged care\" in IEEE Access, 7, 171173- 171194, 2019.  Amershi, S., et al. \"Guidelines for human- ai interaction\" In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, pp. 1-13.  Ankil Shah, Bhargav Jain, Bhavin Agrawal, Saurabh Jain, Simon Shim, “Problem Solving Chat bot for DataStructures”, IEEE, 2018.  Bhaumik Kohli, Tanupriya Choudhury, Shilpi Sharma, Praveen Kumar.,” A Platform for Human- Chat bot Interaction Using Python”, IEEE , 2018.  Tussanai Parthornratt, Pasd Putthapipat, Dollachart Kitsawat , Prapap Koronjaruwat,” A Smart Home Automation via Facebook Chat bot and Raspberry Pi”, IEEE, 2018.  Jitendra Purohi,, Aditya Bagwe, Rishbh Mehta, OjaswiniMangaonkar , Elizabeth George, “Natural Language Processing based Jaro-The Interviewing Chatbot” , IEEE, 2019.  Holzinger, A., Biemann, C., Pattichis, C. S., and Kell, D. B., \"What do we need to build explainable AI systems for the medical domain?\" arXiv:1712.09923, 2017.  Clark, L., et. al., \"What Makes a Good Conversation?: Challenges in Designing Truly Conversational Agents\" In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, p. 475, ACM.  McDuff, D., and Czerwinski, M., \"Designing emotionally sentient agents\" in Communications of the ACM, vol. 61, no. 12, pp. 74-83, 2018.  S. A. Prasetya, A. Erwin, M. Galinium, \" IMPLEMENTING INDONESIAN LANGUAGE CHATBOT FOR ECOMMERCE SITE USING ARTIFICIAL INTELLIGENCE MARKUP LANGUAGE (AIML)\", 2018.  S. Hussain, O. A. Sianaki, N. Ababneh, \"A Survey on Conversational Agents/Chatbots Classification and Design Techniques\", 2019.  A. Mondal, M. Dey, D. Das, S. Nagpal, K. Garda, \"Chatbot: An automated conversation system for the educational domain\", IEEE, 2018.  Falguni Patel, Riya Thakore, Ishita Nandwani, Santosh kumar Bharti, “Combating Depression in Students using an Intelligent Chat Bot: A Cognitive Behavioral Therapy”, IEEE ,2019.
Copyright © 2022 Dr. G. Niranjana , DJNV Sai Trinadh. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Paper Id : IJRASET40014
Publish Date : 2022-01-20
ISSN : 2321-9653
Publisher Name : IJRASET
DOI Link : Click Here