SEC APP: UI/ UX DESIGN

SRISHTI COURSE ALLOCATION APP

Made in Collaboration with Santoshi Balaji

Concept Note

The Srishti Course Allocation system that is predominantly used in the beginning of every academic year, and occasionally during the beginning of new semesters, while extremely thorough and effective, can be confusing and time-consuming for both- faculty and students alike.

 

As the student body grows in size, it becomes harder year-after-year to manage this system- even if it’s just a one-time issue.

Thus, we have come up with the idea of the Srishti’s Course Allocation Agent.

 

The communicative agent can act as a complimentary and perhaps even a subsidiary for helping staff- by communicating with the student to sort, filter and organise their course schedule, as well as recommend appropriate courses on basis of keywords spoken by and the questions asked by the user. It can refer to the student’s past records and ask for

course expectations during the communication in order to give the best permutation and combinations of courses, automatically arranging them according to their time slot, the days that they are held, the major that they are categorised under, and the specialisation that they focus on.

The Problem Statement

Most students spend time reading the descriptions of each course, filter their choice of courses and sit down to create messy diagrams to segregate the best combination of courses.

 

Often a student chooses between courses due to their lack of motivation to read the entire description, or because they’re both open on the same time. This expects one change, and this alters the entire diagram.

 

Faculty have to sit for hours to review each form, find errors, and contact & remedy every error in consultation with the student. Even more time is spent explaining course details, and best combination possible given their past record & future expectation.

What does the system do?

The idea is to create a conversational interface that simplifies student’s as well as faculty’s tasks as follows:

  • Compare courses through years to suggest new courses.

  • Sidelining similar courses to those the student has already completed.

  • Filter courses based on credit count, time or day criteria.

  • Records and Reprimands of the number of courses taken versus the number of courses mandatory.

  • Takes intuitive actions like stopping the show of courses for filled criteria unless asked otherwise.

  • Notify the user of their room number for the allotted classes.

  • Sorts on basis of time, credits, major, etc.

IS CONVERSATIONAL AGENT ACTUALLY HELPING OUR ALLOCATION SYSTEM? HOW?

The knowledge base would be in the form of a Database (made on Spreadsheet & imported to a DBMS- Microsoft Excel and Access respectively) containing all the information about each course including the course name, description, credits awarded, days, time, major/ minor, semester, cycle/ workshop, level, and pre-requisite. As a relational database, it will have a SQL language to manage and update it (physical querying and reporting on server).

A Rule Base would contain programming of dos and don’ts for the agent itself- a set of rules that it needs to follow while communicating with the user, i.e. ‘in case the answer to a question is unknown, reply with “Sorry, I don’t understand”, or “Can you elaborate”.

 

An Inference Engine would have an in-built programming that bridges the gap between the Knowledge and Rule base. When the agent registers an input keyword, the engine references to the knowledge base to filter the course(s) that pertain to the input on basis of said keyword. It then refers to the rule base to deduce how that information should be put forth to the user on the User interface.

The last component, the User Interface is the face of the agent that communicates with the user and bridges the gap between the Inference Engine and the User itself through the designed screen.

While the current agent’s User Interface is predominantly screen-based, we are seeking to further develop a verbal communicative device.

Design and Structure

So far, we have designed a system based on the architecture of Expert Systems used in many business organizations for sorting and decision making purposes, that will comprise of 4 components that work together by referencing, learning and crosschecking from one-another to run the agent.

The Knowledge Base

The Rule Base

The Inference Engine

The User Interface

Final Concept

The communicative agent can act as a complimentary and perhaps even a subsidiary for helping staff- by communicating with the student to sort, filter and organise their course schedule, as well as recommend appropriate courses on basis of keywords spoken by and the questions asked by the user. It can refer to the student’s past records and ask for course expectations during the communication in order to give the best permutation and combinations of courses, automatically arranging them according to their time slot, the days that they are held, the major that they are categorised under, and the specialisation that they focus on.