Skip to content
A machine learning verbal application for CSUSB's main website search engine.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
README.md
delim.cpp
fs_script.js
irb_data.json

README.md

AuoraData

A machine learning verbal application for CSUSB's main website search engine. Aurora Data Bot About Me What the project does: An trained interactive feature for devices that use machine learning algorithms to learn from user questions to answer accurately. Using Dialogflow and Firebase Firestore, once a question is asked, this bot will take that question, analyze it, look for certain words (intents), and match the words to entities that the database holds. Flexibility is the main concern of this project, in order for it to be trained and ready to answer a variety of questions. The goal of having intents is to train the bot be able to answer questions that can come in a variety of ways. The entity matching is when the data in the database is triggered by the words in the question and then looks to respond accurately using a Realtime database.

When the bot responds, it will have searched through the Firestore database and matched up the data and then sent the answer through Dialogflow. If it is an unclear question the bot will ask the user to clarify until it understands what the user is trying to ask. This clarification process is within the limit of the database that the bot was originally updated from.

Why the project is useful: The luxury of using an application with a wide database, is it’s effortless use, fast responsiveness and accurate response to a question it has been asked. This Aurora Data project resembles Google Home, Amazon Alexa or Siri. This feature has been proven useful by the general public as an alternative option of how to use a search engine. This will be added to the California State University, San Bernardino main website and is not only an alternative option used for a hands-free feature, but also as a machine learning feature to automatically update when common questions are asked or when new data has been added to the database. How users can get started with the project: What makes this such a successful project is the ability to show other people exactly how this was accomplished. First, users will want to get familiar with the platforms that they will be working with Dialogflow and Firebase Firestore. Along with learning about the platforms, knowing the difference between SQL and NoSQL and why using a relational database is important to use in this project. Where users can get help with your project: Utilizing as much information on the platforms is important for base knowledge, but not everything has been explored or has not been documented to help to know what the next step may be. Ways that you can get help on this project as you go through it would be to follow the steps that have been documented here. Who maintains and contributes to the project: The organization that will be monitoring this project is the ITS department at CSUSB and the 6 students in the Independent Research course.

Project Overiew & Objective

Dialogflow

Overview

Terminology

Configuring Dialogflow

Writing Fulfillment Code

Firebase

Overview

Setting up Firestore

Afterword

You can’t perform that action at this time.