EC601_Project：Automated Admission Platform

" It's not a bug, it's a feature. " -- Haoyang W.

Haoyang Wang, Qingxuan Pei, Kefan Zhang

Demo

(Web App): https://threehungrychinese.github.io/MaineCoonUi/

(iOS App): (Checkout screenshots of iOS App or contact [email protected] for it) https://github.com/ThreeHungryChinese/iOSApp

Project Repos：

https://github.com/ThreeHungryChinese/EC601_Project Main readme of this prjcect by all group members

https://github.com/ThreeHungryChinese/EC601_Project Regression scripts and server configurations by Kefan Zhang

https://github.com/ThreeHungryChinese/iOSApp iOS Application by Kefan Zhang

https://github.com/ThreeHungryChinese/WebApps [abandoned] Web Application (Sprint 1 & Sprint 2) by Haoyang Wang

https://github.com/ThreeHungryChinese/MaineCoonUi Vue web interface (Sprint 3) by Haoyang Wang

https://github.com/ThreeHungryChinese/MaineCoonApi Back-end services (Sprint 3) by Haoyang Wang

https://github.com/ThreeHungryChinese/PDFprocess PDF Process Algrothim by Qingxuan Pei

Project Guide

Web Application

A user Interface based on Vue.js

features

• Material Design User Interface
• COOL visible design for building a data flow
• Elegant Animations
• High completion level

Demo

Sample User:

For developers: [email protected] | Password:123456

For School Administors: [email protected] | Password:123456

For Students: [email protected] | Password:123456

Demo Link:

https://threehungrychinese.github.io/MaineCoonUi/

Instructions & ScreenShots

For developer

After login the system, UI will look like this

Try it by clicking edit & New & Delete Button

Set up program's data flow

In the Edit/Create dialog's third step, there is something like this

• As you can see, there are bubbles (We call it dots) on users screen. Blue dots represents this Programs' input which get from student's input and the output score of this flow. So this means the input dots must have a pointer to others and the output must be pointed from other dots.

• Other dots means a algorithm provided by developers. Users can click the add button in the top toobar to and a dot or delete button, then click a dot to delete one.

• If a user click a dot, its input setting table will show on the below. And the dot's color will change to orange, which shows this dot is selected.

• Once a user set all the input of a algorithm, it will change to green, which means this dot is set correctly.

• All dots are dragable, so if one are setting a huge data flow, they will find this feature is really useful.

For School Adminisator

After login the system, UI will look like this

Try it by clicking edit & New & Delete Button.

For Student

After login the system, UI will look like this

click start to start a prediction.

Deploy

Dependency

"core-js": "^3.4.8",
"vue": "^2.6.10",
"vue-d3-network": "^0.1.28",
"vue-material": "^1.0.0-beta-11",
"vuebars": "^0.1.5",
"vuelidate": "^0.7.4"

Set up developing Environment

First: Download a Node.js & Environment

Check https://www.npmjs.com/get-npm to get instructions about how to install it on your OS

Second: Donwload Vue-Cli

(Reference https://cli.vuejs.org/)

npm install -g @vue/cli

Third: Launch Vue-Cli

Clone this repo.

To open this project, please just change your current dictionary of your terminal to the repo's root dictionary

And then,

vue ui

Deploy on your own server

need a web server, like Apache || Nginx || IIS.

just link the index.html will not work since browser's CROS policy may block some essential resources.

Important information about CORS

Check to get information about CROS

iOSApp

MaineCoonApi

RESTful API for automation admission system

Based on .Net Core 3.0

Provide services to all comptiable client application

Check a web application using this Api Service

https://github.com/ThreeHungryChinese/MaineCoonUi

Check this projects' readme

https://github.com/ThreeHungryChinese/EC601_Project

Sprint 1

Outcome:

·Product Definition Document(Product Mission, Customer(s), MVP user stories)

·All system setup

·Review and analysis industrial products

·Test programs

Sprint 2

Outcome:

·Extended to 6 default algorithms.

·Developed web platform, including interface, register and login function, feasible to apply third-party algorithms or choose different default algorithms.

·API for Thrid-Party Developer

·HTTPs request to communicate between different components of system

Schedule:

Stage #1:(Sep 20th - Oct 4th)

Assignment: Product Definition Document, System Setup and Test Programs.

Stage #2:(Oct 8th - Nov 4th)

Assignment: System architecture (HTTP request to return score, data storage and export function), more algorithms, explore feasibility of NLP for LOR, SOP (VMOCK), extension on Telegram (*undecided)

System Architecture:

Sprint 3

Outcome:

·Web Application is abandoned. (In sprint 3, it is replaced by high choesion front-end Vue app and .Net Core Restful Api service.)

·Add a new Web client 'MaineCoonUi', with a elegant, Material Design outward, used Vue.

·Add a back-end service 'MaineCoonApi', it is able to provide a Restful api to other front-end application, used .Net Core 3.0.

·Rewrite Apis for developer & update the SQL Server tables

·Developed iOS application for students.

·Added a sample algorithm allow user to convert Transcript to average score.

·Improved sample algorithm.

New System Architecture:

Product Mission

The product is composed of three main parts: the user interface, the admission assistant system and data storage structure.

The user interface, i.e. UI, supports users to upload the required documents for applying. After the result calculated by the admission assistant system is generated, the chance of admission is shown on the user interface.

The admission assistant system is a rating system that takes required documents as input and outputs a potential score as the chance of admission. The system is supported by several different algorithms and regard the admission possibility as a supervised learning problem.

Another feature of this product is that we add a data storage structure. With the help of this, users in the future can acquire admission data and the corresponding documents for statistical use.

Customers

1.Applicants

2.Admission Officers

3.Developers

MVP(Minimum Valuable Product) User Story

1.for applicants

Applicants can use this product to evaluate the chance of him/herself being admitted by some certain university/college.

2.for admission officers

Admission officers can use this product to assist them to comprehensively consider whether an applicant is qualified to be admitted.

3.for developers

Since we will reserve the API interface and corresponding API standard, developers who have developed great evaluating algorithms can test their algorithm reliability via our API. For profit purpose, they can consider selling it to universities if the algorithm performs better than the existing algorithms.

Minimum Valuable Product before Sprint 1:

design a system with:

· an interface that takes in documents and outputs the scores/chance of admission

· an admission assistant system with >=1 algorithm(s) available to generate evaluation scores/chance of admission

· a data storage structure that stores all data