/JS18_ProjectA_Group2

In this project we created the framework that translates natural language to data visualization creation. This project encompasses loading and querying data and creating simple graphs.

Primary LanguageTypeScriptGNU General Public License v3.0GPL-3.0

JS18_ProjectA_Group2

Build Status

Requirements

Your system must have support for below stuff:

  • NodeJS.
  • Python 3.6.
  • MySQL.
  • A Modern Browser.

Set Up

  • Install fresh rasa_nlu instance in the root directory. JS18_ProjectA_Group2

  • Install Rasa_NLU from here

    [ Don"t clone Rasa_NLU original git repo as it contains different version than what we use ]

    Follow below steps:

Database Configuration

  • Install MySql server.

  • Create database with name igraph

  • Import JS18_ProjectA_Group2/server/sql/config.sql into igraph.

  • Provide MySql server credentials in JS18_ProjectA_Group2/server/config.js

Reset Database to original(No dataset)

  • Drop all the tables except config.
  • Delete all the entries from config table
  • Delete all the files from server/data folder

Starting Project

  • Open a terminal on JS18_ProjectA_Group2 folder. Run below command to install all the dependency for both server and ui.

    npm install

  • Run below command to test if basic functionalities are working fine for server and ui.

    npm test

  • Start Mysql server. We use XAMPP PHPMyAdmin to run our mysql server.

  • To run the application execute below command.

    npm start

  • If you want to run server, ui and nlp engine separately and manually, execute below commands one by one in different terminals from the JS18_ProjectA_Group2 folder.

    npm run server

npm run ui

npm run start:nlp

  • If any issue running nlp_engine command mentioned above, run below python script from terminal on JS18_ProjectA_Group2/nlp_engine folder to check the issue.

    python -m rasa_nlu.server -c sample_configs/config_spacy.json

How to use!

  • We don't have pre-uploaded dataset in the system, Please upload dataset file.

  • Download sample kaggle datasets from here and here.

  • Upload sample datasets using the application upload feature. So that necessary features will be processed before using them for plot.

  • We currently don"t support deletion of files uploaded into database. If manually done, make sure to delete corresponding file from server/data and corresponding row from config table.

Environment Variables

For MySQL:

  • DB_HOST - host name or ip address (defaults to localhost)
  • DB_PORT - port number (defaults to 3306)
  • DB_USER - login/user name (defaults to root)
  • DB_PASSWORD - password (defaults to empty password)
  • DB_NAME - database name (defaults to igraph)

Proxy Configuration

Proxy configuration has been done on UI to access nlp engine ('/nlp') and backend ('/api').

Supported Queries

  • CORE_DATASET

    • Plot histogram of employee’s age
    • Plot number of employees for each gender as a pie chart
    • Plot count of employees for each gender as a pie chart for all departments
    • Plot average pay rate of each race as a bar chart
    • Make a scatter plot with age and pay rate
    • plot histogram of number of employees per department
    • Draw bar chart of number of employees of each status
    • Plot count of employees at each position in every department as a bar chart
    • Show me the distribution of age on a histogram
    • Show number of employees for each MaritalDesc whose age is between 28 and 32 on a bar chart
    • Plot a scatter graph of average age vs average salary for each department where average salary is more than 10.
    • Plot a line graph of maximum age vs max salary for each department having highest age is less than 50.
    • Plot a scatter graph of average salary vs average age for each department where average salary is more than 10.
    • Pie chart of average salary per department having mean age greater than 25.
    • average age vs maximum salary on a bar chart per department
    • Show number of employees vs maximum salary on a bar chart per department

  • INDICATOR_DATASET

    • Plot the histogram of “Urban population (% of total)” in the year 2010
    • Plot the pie chart of “CO2 emissions (kt)” for each year
    • Plot the pie chart of “Debt buyback (current US$)” each countryname.
    • Plot the bar chart of “Debt buyback (current US$)” for each countryname per year
    • Plot the bar chart of “External debt stocks (% of GNI)” for each countryname with year lesser than 2000
    • Plot the bar chart of “Debt buyback (current US$)” for each countryname where year is between 2000 and 2010
    • Plot the scatter chart of “GDP per capita (current US$)” vs year for each year
    • Plot the line chart of “Mobile cellular subscriptions” vs year for every year

Project Architecture

Project Architecture

Credits

Logo is taken from Vecteezy

Author