Conversational Interface for Supply Chain Risk Management Data Visualization

How might we make an easy-to-use interface for lay users to explore data in the supply chain? This project aims to solve this problem using a conversational user interface and interactive data visualizations. Several key design decisions are made to achieve the goal. For example, how to match the user interface with data novice's mental models of constructing data visualizations? How to integrate structured and natural language modalities so that users can get a sense of control? How to use conversational queries to disambiguate and simplify the input? Two conversations are designed for the project demo, along with interactive data visualizations coded with Bokeh in Python and D3.js in JavaScript.