Conversation-Centered AI (Chatbot) System

A Conversation-Centered AI (Chatbot) System is an interactive AI-based system that can solve conversation-centered AI tasks (designed to engage in conversations modeled on human-to-human discussion).

• Context:
  • input: Text Chatbot Request.
    • Optional Input: Unstructured Chatbot Input, such as: document file or an image file.
  • output: Text Chatbot Response.
  • measures: ... [[]] ... human-likeness of their conversations.
  • It can (typically) have a Conversation-Centered UI.
  • It can (typically) have Chatbot Features, such as: conversational abilities, accuracy of responses, user satisfaction, and the ability to handle different types of queries effectively.
  • It can (typically) accept GenAI Model Prompts.
  • It can (typically) be associated with a AI Chatbot Initialization Prompt (like ChatGPT initialization prompt).
  • It can (typically) utilize NLP Technologies to enable natural conversational abilities.
  • It can (typically) have a Conversational AI User Interface.
  • It can (typically) be developed through a Chatbot Development Task.
  • It can (often) be described in a Chatbot PRD.
  • It can (often) be developed by a Chatbot Development Team (with chatbot developers).
  • It can (often) be based on a Chatbot Platform (either a custom chatbot development platform or a configuration-based chatbot development platform).
  • It can (often) be an LLM-based Conversational AI Assistant System.
  • ...
  • It can be modeled with a Chatbot System Architecture.
  • It can range from being an Freeform Chatbot to being Structured-Dialog Chatbot.
  • It can range from being an Information-Providing Chatbot System to being an Action-Taking (Agent) Chatbot System.
  • It can range from being a Conversational AI System with Free-Flowing Dynamic Dialog to being a Conversational AI System with Predefined Responses.
  • It can range from being a Open-Topic Conversational AI System to being a Domain-Specific AI Assistant or a Task-Specific Conversational AI-based System (such as a customer service conversational AI system).
  • It can range from being a Data-Driven Conversational AI System to being a Knowledge-Enriched Conversational AI System, based on knowledge sources used.
  • It can range from being a Memoryless Conversational AI System to being a Conversational AI System with Memory.
  • It can range from being a Personalized Conversational AI System to being a Non-Personalized Conversational AI System, based on whether it uses a user context.
  • It can range from being a Public Conversational AI System to being an Enterprise Conversational AI System, based on accessibility.
  • It can range from being a Paid Conversational AI System to being a Free Conversational AI System.
  • It can range from being a 3rd-Party Conversational Chatbot to being a ___ Conversational Chatbot, based on commercial model.
  • It can range from being an Experimental Conversational AI System to being a Production Conversational AI System, based on maturity level.
  • It can range from being a Cloud-Based Conversational AI System to being an On-Device Conversational AI System to being a Hybrid Conversational AI System, based on its technical architecture.
  • ..
  • It can be evaluated by a Chatbot Evaluation System (possibly using a chatbot evaluation dataset).
  • It can create Chatbot Session Log Data.
  • …
• Example(s):
  • a Freeform Casual Conversation Chatbot such as XiaoIce (open-ended chit-chat systems), that engages users in casual conversations about a wide range of topics.
  • a Social Robot such as Jibo (social interaction assistants), that carries on open-ended chit-chat with users to build emotional connections and provide companionship.
  • a Conversational Instruction-Taking Chatbot such as Amazon Alexa (voice assistants), that takes user instructions to perform tasks like playing music or controlling smart home devices.
  • a Domain-Specific Chatbot such as DoNotPay (legal assistants), that assists users in generating legal documents or contesting parking tickets.
  • a Task-Specific Information-Providing Dialog-Centered Chatbot such as Bank of America's Erica (financial assistants), that provides users with banking information, transaction details, and account management assistance.
  • a Task-Specific Dialog-Centered Agent such as Connie the Hilton Robot (hospitality assistants), that helps guests with booking rooms, checking hotel amenities, and answering frequently asked questions.
  • an Action-Taking (Agent) Chatbot such as Google Assistant (smart assistants), that can book a taxi ride or set reminders based on conversational commands.
  • a Multi-Modal Open-Ended Action-Taking Chatbot such as OpenAI's ChatGPT with Plugins (multi-functional assistants), that supports a wide range of conversation topics and actions, including integrating with third-party services through plugins.
  • a Conversational Personal-Assistant such as Siri (virtual assistants), that performs various tasks including managing calendars, sending texts, and setting alarms.
  • an Conversational Enterprise Search System such as IBM Watson Assistant (enterprise assistants), that is queried by employees to find documents, data, and answers within an organization.
  • a Data Analytics Conversational AI System such as Yellowfin's Signal Chatbot (data analytics assistants), that generates insights and data visualizations from user queries.
  • an Application Copilot Chatbot such as GitHub Copilot (coding assistants), that assists developers by providing code suggestions and completing functions in real-time as they write code.
  • a LLM-based Chatbot such as Google Bard (conversational assistants), that provides users with detailed information and creative content through natural language conversation.
  • a Customer Service AI Assistant such as Zendesk's Answer Bot (customer support assistants), that resolves customer inquiries and automates common support tasks, improving efficiency and customer satisfaction.
  • a Healthcare Conversational AI System such as Nuance's Dragon Medical Assistant (medical assistants), that helps healthcare providers document patient interactions, manage records, and enhance the accuracy of clinical documentation.
  • a Sales Assistant AI Chatbot such as HubSpot's Chatbot Builder (sales assistants), that interacts with potential leads, qualifies them, and directs them to the appropriate sales channels.
  • a Job Interview AI Chatbot such as HireVue (recruitment assistants), that conducts initial candidate screenings and provides recruiters with insights on potential hires.
  • an Educational AI Chatbot such as Quizlet's Learning Assistant (education assistants), that helps students study by generating practice quizzes, answering questions, and explaining concepts in a conversational manner.
  • …
• Counter-Example(s):
  • A Rule-based Conversational System, such as ...
  • A Menu-based AI-based System, such as Grammatical Correction System.
  • A Wizard-based System that does not accept conversational input), such as a wizard-based personal tax analysis system (such as TurboTax).
  • A Form-based System, such as a e-commerce checkout form.
  • a Domain-Specific Information-Providing System such as a weather chatbot that cannot carry a conversation.
• See: Conversational AI, Conversational Agent, Intelligent Virtual Assistant, Intaractive AI-based Application, Chatbot API Design, Generative Artificial Intelligence, Software Agent, Chatbot Analytics.

---

---

References

2023

• (Wikipedia, 2023) ⇒ https://en.wikipedia.org/wiki/chatbot Retrieved:2023-11-21.
  • A chatbot (originally chatterbot) is a software application or web interface that aims to mimic human conversation through text or voice interactions. Modern chatbots are typically online and use generative artificial intelligence systems that are capable of maintaining a conversation with a user in natural language and simulating the way a human would behave as a conversational partner. Such technologies often utilize aspects of deep learning and natural language processing, but more simplistic chatbots have been around for decades prior.

    As of 2022, the field has gained widespread attention due to the popularity of OpenAI's ChatGPT (using GPT-3 or GPT-4), released in 2022, followed by alternatives such as Microsoft's Bing Chat (which uses OpenAI's GPT-4) and Google's Bard. Such examples reflect the recent practice of such products being built based upon broad foundational large language models that get fine-tuned so as to target specific tasks or applications (i.e. simulating human conversation, in the case of chatbots). Chatbots can also be designed or customized to further target even more specific situations and/or particular subject-matter domains.

    A major area where chatbots have long been used is in customer service and support, such as with various sorts of virtual assistants. Companies spanning various industries have begun using the latest generative artificial intelligence technologies to power more advanced developments in such areas.

2023

• (IBM, 2023) => https://www.ibm.com/topics/conversational-ai
  • QUOTE: ... Conversational AI combines natural language processing (NLP) with machine learning. These NLP processes flow into a constant feedback loop with machine learning processes to continuously improve the AI algorithms."

2023

• (Google Cloud, 2023) => https://cloud.google.com/conversational-ai
  • QUOTE: ... Conversational AI works by using a combination of natural language processing (NLP) and machine learning (ML). Conversational AI systems are trained on large amounts of data, such as text and speech. ...

2023

• (TechTarget, 2023) => https://www.techtarget.com/searchenterpriseai/definition/conversational-AI
  • QUOTE: ... Natural language processing (NLP) is the current method of analyzing language with the help of machine learning used in conversational AI. ...

---