IndiOne : Code-with-Google-Maps

A mobile-friendly web application that automatically schedules itinerary based on user preferences

demo.mp4

About Team

• Team name: IndiOne
• Members
  • (Leader) Minseok Kim - [email protected] - Backend Dev
  • Siwoo Jung - [email protected] - Frontend Dev
  • Seonghyun Kim - [email protected]

Problem Statement

As AI technologies have been developed and made available to the public over the last decade, the demand for a convenient lifestyle has become a growing global issue.

We organized our ideas and devised a plan to address this trend. As a result, we have created an application with the function of recommending multiple routes for customers with various purposes (dating, leisure trips and business, etc.).

Selectable options are integrated into the system to allow users to customize their plans accurately, according to their preferences.

A Brief of the Prototype

UML

Prototype sample images

• This is the main page where a user can input their choices of

  

  • Travel mode

  • Starting location

    • New swipeable edge pops up

    • Search the location with Place Autocomplete

      

  • Travelling time

  • Maximum distance to travel

  • Budget level

    • This level corresponds to 'price_level' from Places API

  • Transportation mode

• Results

  • Recommended routes with four different themes - casual, local specialty, qualitative and shortest - are shown

    
    • Casual - the route has the highest score in overal.
    • Local Specialty - the route include the tourist spot.
    • Qualitative - the route has the highest rating of the places.
    • Shortest - the route has minimum moving range.

  • A user can navigate recommended routes of different themes with the BACK and NEXT buttons at the bottom

  • A user can open each itinerary card to view detailed descriptions

    

Tech Stack

Backend

• Backend framworks
  • Python Flask
• Database
  • Mongo DB

Frontend

• UI
  • JavaScript, HTML, CSS
• Frontend frameworks
  • React.js
• Styling
  • Bootstrap, MUI
• Web application frameworks
  • Node.js

Instructions

Backend

• Create and activate Venv

  • Move to Backend folder in your terminal
  • Type python -m venv BackendVenv (for mac: python3 -m venv BackendVenv)
  • Type ./BackendVenv/Scripts/Activate (for mac: source BackendVenv/bin/activate)
  • Select the python.exe in the BackendVenv/Scripts folder as interpretor of VSC (for mac: python.exe is in the BackendVenv/bin folder)

• Install packages

  • Type pip install -r requirements.txt

• Input API key

  • Create a json file call keys.json
  • Paste your api key in the file

    {
      "googleApi": "GOOGLE API KEY",
      "databaseUrl": "MONGO DB API KEY"
    }
    

• Run Server

  • Run server.py in the src folder
  • If following message appears at the bottom of the terminal, the server is ready

    * Serving Flask app 'IndiOneApi'
    * Debug mode: off
    WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead.
    * Running on http://127.0.0.1:5000
    Press CTRL+C to quit
    

• Deactivate venv

  • Type deactivate

Frontend

• Manage API

  • Locate 'Frontend' folder
    • cd Frontend
  • Create .env file
  • In .env file, insert the following line, with "YOUR_API_KEY" replaced by your API

    VITE_API_KEY=YOUR_API_KEY
    

  • For example,

    VITE_API_KEY=A1B2C3D4E5F6G7
    

• Install packages

  • Type npm install --legacy-peer-deps
  • Make sure to include --legacy-peer-deps

• Launch application

  • Type npm run dev
  • If your web browser does not automatically launch, locate addresses shown in the terminal
  • For example,

    Local:   http://localhost:1234/
    

• Run application

  • Choose your pereferences
    • Choose Travel Mode, one of 'Solo', 'Couple', 'Friends' or 'Family'
    • Choose your starting location
      • Locate From where? and click/tap the search bar
      • A new swipeable edge is shown. Click/tap 'Starting Location' bar and input your starting location
      • Close the swipeable edge
    • Choose your travelling time
      • Pick your start and finish times
        • Finish time shall not be earlier than starting time
        • Start time + finish time cannot be more than 24 hours
    • Choose maxium distance you would like to travel
      • Locate the slide bar and choose the distance between 1 ~ 50km
    • Choose your budget
      • Choose a budget level between 0 and 4 ('$' is level 0)
    • Choose your *transportation mode
    • Press 'FIND MY ITINERARY' button at the botom of the page
  • Wait until the result page is shown

• View results

  • Once the result page is loaded, it shows four different routes based on the themes of casual, Local Specialty, Qualitative and Shortest
    • A user can navigate with BACK or NEXT buttons at the bototm of the page
  • A user can open or close a description card below the map to view detailed information of each itinerary
  • Press HOME icon at the bottom to return to the initial page

Future Scope

In the future, following improvements could be applied

• Improved route recommending algorithms
  • Apply machine-learning for users' route choices and apply it for route recommendation algorithms
  • A feature that a user can choose unwanted places among recommended routes, and those unwanted places to be replaced by other places while maintaining the rest of the places
• Handling user database
  • Sign-up, login, logout features
  • Saved user preferences
• Overall improved UI
  • More detailed description of each selected places
    • Sample photos
    • Link to the official site
• Customization of Google Map - simple, yet informative