JARVIS for Corporate

• Problem Statement Number : SM446
• Organization : FIS Global

K3RNEL-PAN1C.ASEC (Team Members)

• Atharva Veer (Team Leader)
• Yash Jain
• Harsh Chheda
• Kritika Ravishankar
• Dhwani Agarwal
• Divy Patel

Technology Stack

• SNIPS NLU for intent parsing in chatbot
• Flask for server-side development
• Spacy for extracting MoM entities from meeting texts
• HTML-CSS-JS / Bootstrap Web technologies for building user interface
• PyDub for chunking required for Minutes of the Meetings
• [email protected] for sending E-Mails

Application Architecture

Features implemented

• User-interface to interact with chatbot via Speech generated or raw text user queries
• User-interface can display response of the query on the screen or can send via email
• Supports recording the meeting, summarizing and getting the appropriate details of the meeting from database depending on the keyword
• Supports Stack-Overflow queries
• Supports queries for Atlassian products:
  • Jira
  • Confluence
  • Bitbucket
• Supports aggregated and comprehended information retrieval through query

Key Characteristics on Performance and Security metrics

• Chatbot and application hosted on a server in the intranet
• No third party server interference
• Flexible code, easily pluginable interface
• Ability to identify closest queries even in case of mis-spelled utterances

Working of Chatbot

Comparison between various Chatbots

Installation and Setup

Prerequisite

    Python 3.7
    Local Instance of Jira, Bitbucket, Confluence running

Command to install the Python package dependencies

$ pip install -r requirements.txt

Command to install en(English) language support for SNIPS-NLU intent parser

$ python -m snips_nlu download en

Command to install Spacy's en(English) corpus

$ python -m spacy download en_core_web_sm

Train and load the chatbot model (Powershell script)

$ cd ./NLU/scripts
$ .\train_engine.ps1

Starting the server

$ python app.py

Application localhost URL

http://localhost:5000/

References

• Flask Official Documentation : https://flask.palletsprojects.com/en/1.1.x/
• SNIPS NLU Official Documentation : https://snips-nlu.readthedocs.io/en/latest/
• Spacy Official Documentation : https://spacy.io/models/en
• PyDub Official package source : https://pypi.org/project/pydub/
• Jira Server API Documentation : https://docs.atlassian.com/software/jira/docs/api/REST/7.6.1/
• Confluence Server API Documentation : https://developer.atlassian.com/server/confluence/confluence-rest-api-examples/
• Bitbucket Server API Documentation : https://developer.atlassian.com/server/bitbucket/reference/rest-api/