hotelchatbot/chatbot_for_hotel

Intelligent ChatBot for Better Hotel Experience

Online URL: Hotel Chatbot

Basic Requirement

Local deployment only supported on Mac OS X, might not work for other operating systems.

Installation

1. Download the project

git clone https://github.com/HotelChatbot/ChatBot_for_Hotel.git

2. Enter the project directory

cd ChatBot_for_Hotel

3. Download all the dependencies

npm i

4. Open another console to activate the database(if haven't)

mongod

5. Go back to original console and load the datacase documents

mongoimport -d hotel-chatbot -c restaurant --type csv --file public/restaurant.csv --headerline --upsertFields name

mongoimport -d hotel-chatbot -c hotel_facility --type csv --file public/hotel_facility.csv --headerline --upsertFields name

mongoimport -d hotel-chatbot -c room_facility --type csv --file public/room_facility.csv --headerline --upsertFields name

mongoimport -d hotel-chatbot -c tourist_spot --type csv --file public/tourist_spot.csv --headerline --upsertFields name

Activation

1. Activate the database(if haven't)

mongod

2. Open another console and start off the server

npm start

3. Open a browser and visit the URL

http://localhost:8080/

Technical Documents

There are two sets of technical document, which are located at public/out_backend/global.html and public/out_frontend/global.html

Simple tips for using

1. The blue microphone button at the left-bottom of your browser is clickable to enable/disable speech recognizer, please click to activate the button when speaking. Feel free to disable it when the chatbot is idle.
2. The message textbox at the left-middle of your browser allows text input to replace speaking if you like. It provides alternative availability for user who is under the circumstance that is not speak-convenient.
3. The mode button at the left-middle of your browser is for developer to test their server availability. Please be reminded that it is not for user intention.

External Systems

There are three external systems build alongside with the core system, including room display, staff screen and restaurant management. The links below support direct access to the aforementioned system:

1. Room Display: link, github
2. Staff Screen: link, github
3. Restaurant Management: link, github

chatbot_for_hotel's Issues

ECONNRESET: D3.js TCP socket layer connection error

The TCP connection from server to the CSV file can break down abruptly, which leads to the failure of csv file loading.

The error message reproduce code:

function read_csv(){
    d3.csv("restaurant.csv", function(err, data) {
        if(data){
            //...
        }

        if(err){
            console.log(err)
        }
    });
}