Sharia Math Bot

Sofronia SJ Lin
4 min readMay 12, 2020

--

Concept

Sharia Math Bot is an educational chatbot built on SMS that can help middle school and high school students solve math problems and learn about social justice issues embedded in the problems. This is a work of creative project for the course 49714 Programming for Online Prototypes at Carnegie Mellon University.

The concept of intelligent tutoring systems (i.e. An intelligent tutoring system is a computer system that aims to provide immediate and customized instruction or feedback to learners, usually without requiring intervention from a human teacher (‘Intelligent tutoring system’ Wikipedia. Available at: https://en.wikipedia.org/wiki/Special_relativity ) has become more and more prevalent in the educational field; however, as breaking down problems involves a lot of steps, the current intelligent tutoring systems have complex interfaces that are not user-friendly. Since chatbot has a a concise interface that is easy to interact with and it is more accessible than ITS on desktop, the author came up with the idea of transferring the intelligent tutoring system to chatbot.

Design

The chatbot is called Sharia, and it was designed to have a personality that mimics a kind and patient math teacher in real life.

Chatbot personality sheet

The bot will provide user with a daily math challenge, and give step-by-step guidance in solving the problem. Here is a workflow diagram for the general process.

These are some basic functions that the bot has. The author used the functionality matrix to prioritize the features that the bot should carry.

Functionality matrix

Prototype

The students will be given a word problem first: for example:

Chatbot providing math problems

And then they will be asked to solve the problem. If a student has difficulties in solving the problem, the bot will give a step by step walkthrough of the problem-solving strategies. User will be prompted to answer some multiple choice questions and receive instant feedback at each step.

Example of problem breakdown
Example of problem breakdown

At the end, the chatbot will give the learner a recap of the problem-solving processes

Process recap

Then, it will ask learners whether they want to learn more about social justice issues and provide some background information about social justice issues that are relevant to the problem, such as gender and race pay gap.

Social justice information

Reflections

While developing the prototype, the author encountered issues such as requirement of large amount of user input. To solve this problem, she turned the user-input prompts from open-ended questions into multiple-choice questions. The trade-off of this measure is that user will have less flexibility for inputing their answers.

In addition, currently the bot is less adaptive to user’s responses and is constantly sending messages without leaving enough time for users to process the information. To improve on this, the author might need to use timer to let the bot wait a bit longer between messages.

Also, now the math problems and feedback are generated by human and require a lot of efforts. To make the chatbot more powerful and really adaptive to user’s responses, we need to consider incorporating machine learning and artificial intelligence, which is much more complicated but opens more possibilities for the idea.

The involvement of parents in the process is also a factor to consider, parents may use the chatbot to keep track of the learner’s learning progress and we can introduce a feature to allow parents to choose the math challenges for students,

As it is difficult to have access to target learners, this prototype needs to be further tested and iterated.

Looking forward, the author will do more user testings and confirm the feasibility of the idea. Also, she will improve the prototype by incorporating a timer as well as considering other possibilities such as machine learning and parent involvement.

--

--