A agent, or Artificial Conversational Entity) is a

A chatbot (also known as a Talkbot, chatterbot, Bot, chatterbox, IM bot, interactive agent, or Artificial Conversational Entity) is a computer program which conducts a conversation via auditory or textual methods.Such programs are often designed to convincingly simulate how a human would behave as a conversational partner (1). It is powered by artificial intelligence techniques and natural language processing, image and video processing, and audio analysis to extract contextual meaning and user intent, then respond with a human-like intelligence.

The purpose of the project is the analyze the evolution, modern trends, diversity and capability of chatbots with relevant use cases.The Scope of the project will try to cover and demonstrate the following key aspects of (and not restricted to) the Chatbots with appropriate flows•    NLP (Natural Language Processing – #DM)•    Security and consensus (#SA)•    Context Aware (#AI)•    Self-aware and learning (#AI)•    Simple UI/UX and Personality•    Graceful fallback?LIST OF ABBREVIATION1.    AI- Artificial Intelligence2.    CC – Cloud Computing 3.    DM – Data Mining4.    SA – Software Architecture 5.    VES – Verizon Enterprise Solutions6.    VEC – Verizon Enterprise Center7.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

    POCs – Primary Contacts8.    IdP – Identity Provider9.    SSO – Single Sign-on10.    SAML – Security Assertion Markup Language11.    OAAAS – OAuth Authorization as a ServiceLIST OF FIGURES1.

    VEC Architecture for Chat clients2.    VEC Architecture for Portal integrated bot3.    Slack – VecTestBot4.    Telegram – VecTestBot5.    Skype – VecTestBot?•    Chapter 3: Expected Benefits on the implementation•    Customer Experience: a.    Enhancing the customer perception of VES as a technology leaderb.    Industry recognition via JD Power, TM Forum, positioning in Gartner’s magic quadrant, Root Metrics etc. c.

    Improve OSAT score (Overall Satisfaction score) for online portal•    Call center cost reduction:a.    Initial expected projection of 10% reduction on incoming calls for Call centersb.    On successful addition of use cases time to time will have additional year on year cost savings•    Live User chat cost reduction:a.    Initial expected projection of 20% reduction on Live chat costsThese benefits are calculated based on the value that will be delivered and realized in the next 2 to 3 years.•    Chapter 4: Scope of Work and ObjectivesThe Scope of the project will try to cover and demonstrate the following key aspects of (and not restricted to) the Chatbots with appropriate flows•    NLP (Natural Language Processing – #DM)•    Security and consensus (#SA)•    Context Aware (#AI)•    Self-aware and learning (#AI)•    Simple UI/UX and Personality•    Graceful fallbackObjective of the project is to implement relevant chatbots supporting the above aspects.

Proof of concepts where actual implementations are not deployable in the VEC systems. Figure 1. Message flow for Integrated VEC botsThe Logged in scenario has a live session associated with the BOT and it is easier to validate any request from the Authentication and Authorization perspective.

The Chat window posts all the texts to the BOT Listener. Bot listener passes the query to NLP engine and obtains the Intents’ action. Based on the action, the appropriate VEC API calls are made. •    Chapter 7: VEC Portal External Chat clients Figure 2.

Message flow for Chat clientso    Flow of events:1.    Any chat clients will be associated with OAuth based authentication engine. 2.    Users, after authentication, will be provided with a security token by which all consecutive calls happen.  3.    The chat clients Telegram, Skype and Slack are integrated in the implementation.

4.    Each of the Chat clients will be associated with a NodeJS/Java based Listener, which polls up to the chat clients for messages and passes them to the Engine. 5.    The Listeners looks for the type of the message, and we look forward to support only the Text messages and Voice messages. 6.    Voice Messages: a.    Any voice message will be parsed by the Listener.

Most of the chat clients sends the voice message in OGG format (with OPUS encoding)b.    The Listeners then converts (if needed) the voice messages using the Speech to Text conversion engine and takes the most confident matched text. This text is passed on to the next step.7.    Text Message:a.    The Text messages are validated for any Sensitive inputs and are passed on to the Masking engine.

b.    The Masking engine processes the text iteratively and masks the sensitive data. c.    The Masked data is sent to the NLP engine for actions. 8.    The actions are obtained from the NLP. Based on the Actions appropriate VEC APIs are invoked9.    Based on the response from API, the output for the clients are formatted and returned.

?•    Chapter 8: ScreenshotsFigure 3. Slack Bot  Figure 4. Telegram Bot Figure 5. Skype Bot?•    Chapter 9: Impediments / Challenges and TODOso    Infrastructural challenges?    VEC haven’t provided security clearance for using External Chat clients?    VEC Access Manager doesn’t allow OAuth / SAML based authentication for unknows apps•    No Access to VEC API and data outside of VEC ecosystem•    APIGEE is not exposed to unknow apps/clients as of now?    VEC data cannot be flown through the third-party servers – say for Telegram/Slack etc.o    Technical Challenges?    Challenges on Voice based message •    OGG with OPUS decoder used for most of the chat clientso    OGG _OPUS is a lossy audio coding format for low-latency enough for real-time interactive communication and low-complexity enough for low-end ARM3 processors•    But most of the Voice to Text services support only MP3/WAV/FLAC with various Bit rates•    Speech to Text agents are efficient for Native language speakers (apart from English)?    Amazon Alexa Skill registry needs their device ID to get onboarded?    Bot Approval process, Cost of hosting Applications and for External services•    Bibliography / Referenceso    http://www.

zillman.us/subject-tracers/bot-intelligent-agent-research-resources/o    https://www.cognizant.

com/whitepapers/the-chatbot-imperative-intelligence-personalization-and-utilitarian-design-codex2469.pdfo    https://chatbotslife.com/tagged/artificial-intelligenceo    https://blog.ubisend.com/discover-chatbots/characteristics-best-ai-chatboto    https://www.marutitech.com/heres-need-know-chatbots/o    https://chatbotsmagazine.com/o    https://medium.

com/@jrodthoughts/imagining-a-multi-platform-bot-development-framework-1be4b93f6aafo    https://blogs.oracle.com/dcarru/sp-vs-idp-initiated-ssoo    http://www.bubblecode.net/en/2016/01/22/understanding-oauth2/o    1.

(n.d.). Wikipedia. Wikipedia.

Retrieved from https://en.wikipedia.org/wiki/Chatbot


I'm Casey!

Would you like to get a custom essay? How about receiving a customized one?

Check it out