WAAM
Nolan Windham, Mehrezat Abbas, Abizer Mamnoon, Alejandro Wang
The Premise
WAAM is an AI STEM tutor based on the GPT-4 API. The purpose of this virtual tutor is to guide students in understanding problems on their own. Instead of giving students the answer to a question outright, WAAM encourages users to actively engage in the problem-solving process!
WAAM’s ability to produce relevant, unique visualizations sets it apart from other learning tools and enhances its compatibility with all types of learning styles.
The Goal (A Conversation Mock-Up)
Student: Hi! I'm learning about normal distributions in school. I have a homework problem about data with a mean of 76 and a standard deviation of 10. In a group of 230 data points, how many are above 96?
WAAM: Great! So, you have a set of data that has a mean (average) of 76 and a standard deviation of 10. This means that the values in the set are spread out, but most of them are around 76. One way we can approach this problem is by using the empirical rule. The empirical rule states that for a normal distribution, approximately 68% of the data is within one standard deviation of the mean, approximately 95% with two standard deviations, and approximately 99.7% falls within three standard deviations. How many standard deviations away is 96 from 76?
Student: That would be 96 - 76 = 20, and 20/10 = 2 standard deviations away!
WAAM: Exactly! We know that approximately 95% of the data falls within two standard deviations of the mean. So, the area outside of two standard deviations (which is the area we're interested in) is approximately:
100% - 95% + 0.3% = 5.7%
--NORMAL DISTRIBUTION GRAPH--
Student: Oh! But we only want the data above 96, so we would just want 2.35%, right?
WAAM: Right! So what would the answer be?
Student: 230 x 0.0235 = 5.405
WAAM: Correct! You've got it. Is there anything else I can help you with?
The Goal (An Interface Mock-Up)
The Tools
WAAM was made using Streamlit, GPT-4, LangChain, and JSON
Credit to these sources for helping us out:
https://www.youtube.com/watch?v=ih9PBGVVOO4
https://www.youtube.com/watch?v=TLf90ipMzfE
https://python.langchain.com/en/latest/getting_started/getting_started.html
The Features
The finished WAAM can:
WAAM In Action
WAAM In Action
In case of a reluctant student, it’s important for WAAM to remain encouraging and to adhere to its no-direct-answer policy.
WAAM In Action
Debugging help–WAAM is a virtual grutor!
def in1to10(n, outside_mode):
if outside_mode == True:
if n in (2, 3, 4, 5, 6, 7, 8, 9):
return True
else:
if n not in (1, 2, 3, 4, 5, 6, 7, 8, 9, 10):
return True
return False
WAAM In Action
Moving Forward
Some ways we think this service could be improved even further are:
The Application
A polished WAAM is a valuable tool for organizations and for students. Its ability to fulfill the needs of any student at any level for a low cost provides the opportunity for all learners to get the help they need. The QCL or QSC could use WAAM on their websites to provide students with assistance during off-hours. Schools could provide WAAM to students as an accessible alternative to cheating under duress.
WAAM provides every individual with their own personal super-tutor. Restrictions on direct-answer access and an encouraging demeanor promote deeper understanding of material and confidence in learning universally. No matter what, WAAM will work with students step-by-step in order to provide not only a solution, but a foundation for the future.