Wittwise
Young learners today have inadequate support and have become ultimately in charge of their own learning. For many subject areas, there are actually great resources available online (eg. Khan's Academy videos on Youtube). The problem is that many young learners do not know how to use these resources effectively, resulting in wasted effort and time.
Wittwise is an intelligent software that uses evidence based learning methods to help students learn more effectively. For example, Karpike et al. (2011) showed students who practiced retrieval based learning remember 50% more material than other methods.
The software can analyze a video and produce relevant questions for the student to practice on. Depending on the subject, there could also be practical questions (eg. writing code for an computer algorithm) and project suggestions.
If this solution is scaled globally, it could equalize opportunity for students to effectively learn anything available online.
Students attempting to learn a subject often have to supplement their material with online resources due to different reasons (eg. lack of in-person support). GoGuardian's 2020 analysis (over 30 US school districts) of student online usage found that students frequented educational websites, with Youtube being cited as the students' most preferred source of information. Their students' usage of Youtube had increased by 2x after school closure in 2020.
From Ahn and McEachin's 2017 study of 1.7 million K-12 students in Ohio, they found that students in online schools generally performed less well than traditional in-person schools. They found that it is very difficult for students to manage their own learning process independently, and this is one of the main factors that contributes to online school students having a lower academic score than students attending in-person classes.
In summary, students learning online have the problem of managing their own learning process. The Wittwise software will help solve this problem by automatically providing effective retrieval based methods for secondary students to help improve their retention and comprehension of material.
Wittwise is a chrome browser extension. After a student watches a video (eg. on Youtube), the student clicks on the browser extension and the Wittwise software will analyze the video using AI techniques in natural language processing. Depending on the subject, the student may receive a mix of receive questions to practice retrieval based learning (and suggested answers), practice exercises, and suggestions for experiential projects.
For example, if a secondary student is trying to learn about fermentation in a biology class, the student will receive questions about the fermentation video (along with suggested answers), and a resource for a practical experiment on fermenting a vegetable.
After the student finishes practicing the questions/exercises, the student has the option of publishing it onto their own blog as a portfolio of learning.
The target population are students who use online resources like Youtube as their primary or secondary source of educational information. I am a participant in online communities of students. From trying to keep up to date with the messages, I've noticed that many people are asking questions along the lines of "what is the best resource for X", and constantly jumping from resource to resource. They fall under the trap of the nirvana fallacy, believing that there is a single learning resource that will help them with everything.
I've interviewed some effective students and not so effective students from the online communities, and found that the difference is not in the resource but their learning methods. Both groups use similar resources, but the ineffective learners use the resource passively while the effective learners use active methods (eg. retrieval based methods like flash cards).
I have a small group of eager students who are willing to test the software as it is being released. I will be working with them to continue iterating the software to meet their educational needs. Our measure of success is their academic test scores.
- Increase the engagement of learners in remote, hybrid, and physical environments, including strategies and tools for parental support, peer interaction, and guided independent work.
In remote and hybrid learning environments, students have to be more independent in their learning. As Ahn and McEachin's 2017 analysis of e-school K-12 learners has shown, online learners have poorer results compared to traditional schools. One of the primary reasons is that they do not manage their learning process well since online learning generally has lower engagement with more independent work.
The Wittwise software is a tool to improve engagement with guided independent learning. Specifically, retrieval based practice will be used, which has been shown to be very effective in increasing retention of material.
- Prototype: A venture or organization building and testing its product, service, or business model.
The prototype is currently functional if you take a look at the posted video. Within 2 weeks of the Challenge's deadline, it will be submitted to the Chrome Extension web store and ready to be used/tested.
- A new business model or process that relies on technology to be successful
Many current online learning platforms and instructors focus on providing educational content. They either ignore or provide minimal follow-up for students to active learn from the content. This is because it's expensive to spend more effort producing these materials (eg. practice questions) and many consider it not worthwhile.
Wittwise software can provide the necessary practice questions to help students actively learn and improve their comprehension and retention of material. A software solution like this does not exist in online learning platforms. It has the potential to improve the learning experience of all students using these platforms.
- Artificial Intelligence / Machine Learning
- Crowd Sourced Service / Social Networks
- Software and Mobile Applications
- Children & Adolescents
- Low-Income
- 4. Quality Education
- Not registered as any organization
1 person (part-time)
One time when I was studying a computer science video course on Youtube, I found that there were assignment and practice questions provided by the associated university instructor. Going through those practice questions helped me understand and retain the information much better compared to a previous course that did not have any of these engagement material.
I understood firsthand how different learning could be when it was engaging and interactive with practice questions. But as a learner, I found it difficult to manage my own learning process and create the practice questions myself. That's how the Wittwise solution came about.
As a lifelong learner, I can relate to learners of any age. However, I know secondary school learners do have unique constraints. So that is why I will be working with a small sample of secondary school students to get their input throughout the design and testing process.
- Individual consumers or stakeholders (B2C)
I am applying to Solve because of the valuable partnerships available through the MIT-backed social impact community. Partnerships such as these could help accelerate my work, validate the business model and scale the solution.
Specific barriers include business model validation and scaling to a larger group of people. The MIT-backed community has a diverse range of people, from previous Solve winners to veteran social impact entrepreneurs that can provide good feedback on my business model.
Scaling is also something I do not have as much experience in, and it would be good to have a global community to provide perspective on these challenges.
- Business model (e.g. product-market fit, strategy & development)
- Monitoring & Evaluation (e.g. collecting/using data, measuring impact)
- Product / Service Distribution (e.g. expanding client base)
One of the most important part of an early venture is getting to product-market fit. Since I don't have much experience here, getting mentorship would be valuable. Some aspects around this where I would appreciate advice include: value hypothesis, growth hypothesis, revenue model, target market, service distribution.
As for evaluating the solution's impact, it would be nice to have feedback on the design of the evaluation and data collected.
Century Tech would be a valuable partner organization. They have similar goals in using AI to personalize and advance education. They would be able to help with factors around product-market fit and technology assessment.
- No, I do not wish to be considered for this prize, even if the prize funder is specifically interested in my solution
- No, I do not wish to be considered for this prize, even if the prize funder is specifically interested in my solution
- No, I do not wish to be considered for this prize, even if the prize funder is specifically interested in my solution
- No, I do not wish to be considered for this prize, even if the prize funder is specifically interested in my solution
- Yes, I wish to apply for this prize
The solution is qualified for the AI for Humanity Prize because the core solution uses AI/ML technology and the goal is to equalize access to effective learning methods for online learners.
On the technology side, the solution uses advanced transformers based natural language processing (NLP) models and unsupervised NLP methods.
Because more and more education is moving online, students will have a much more difficult time managing their learning process independently. The Wittwise solution's goal is to help students build these effective learning methods to help them retain information better.
- No, I do not wish to be considered for this prize, even if the prize funder is specifically interested in my solution