Download as pdf or txt
Download as pdf or txt
You are on page 1of 2

Proposal - Bangkit 2023 Company-based Capstone

Company : Dicoding Indonesia


Problem Code : [C23-MP] Mentoring Platform

Team Member:
● (ML) M360DSX2771 - Girvan Syawal Khresnatendi Kurniawan
● (ML) M360DSX1556 - Muhammad Daffa Ferdiansyah
● (MD) A360DSX3046 - Fahreza Rizkia Azhar
● (MD) A071DKX4027 - Kholan Mustaqim
● (CC) C360DSX0863 - Aulia Riefqi Ardana
● (CC) C309DSX0698 - Yusuf Bachtiar

Why is this problem/project interesting for your team?


We want to contribute to Dicoding as they raise the IT educational standards for many learners
like us.

What’s your group’s initial idea to work on this project?


Our idea consists of implementing a mentor matchmaking model not only matching by
standard filtering but also reducing the time cost of matchmaking by predicting mentor’s
willingness and mentee’s acceptance. Based on our research, those concepts have been used
for large coding mentoring platforms like Codementor. We compared this concept with
reciprocal recommendation and hybrid filtering and we found that our concept is the most
suitable solution for mentor-spotting case. We also provide essential features like mentee’s
feedback analytics with sentiment analysis, scheduling, messaging, calendar, and we're also
trying to make our own video call platform.

Does your team have unique solutions to be proposed?


We propose an efficient matchmaking model using two supervised learning models which is
Mentor Willingness Prediction (NWP) and Mentee Acceptance Prediction (MAP).
Collaboratively, those two trained models will find a match between mentor and mentee in the
most accurate and efficient way. We propose four categories of predictive features to
recommend mentors: availability, capability, activity, and proximity. Other than that, we also
provide the best possible all-in-one feedback analytics dashboard for mentors using Natural
Languange Processing (NLP) for feedback sentiment analysis.
Proposal - Bangkit 2023 Company-based Capstone
Based on your knowledge, what tools/IDE/Library will your team use to solve the
problem?

Mobile Development Division: Cloud Computing Division: Machine Learning Division:


● Android Studio (IDE) ● Cloud Storage ● Tensorflow
● Figma ● Visual Studio Code ● Jupyter Notebook
● Jetpack/AndroidX ● Node.js ● Pandas
● Kotlinx ● Hapi ● Scikit-Learn
● Retrofit ● Firebase ● Hugging Face
● Room ● App Cloud ● Google Colab
● JUnit4, Mockito
● Jitsi Meet SDK
● Getstream Chat SDK

Based on your knowledge and explorations, what will your team need support for?
Datasets example for better insight and mentorship to guide us throughout the project. Also
we need Google Cloud credit for the services.

Any other notes/remarks we should consider on your team’s application


It would be an honor to help Dicoding improve it's learning method. Our team consists of
amateur-professional members that both act and learn in this project. Our ML members are
very specialized at NLP so we’re very interested in sentiment analysis. Our MD members are
capable and experienced in mobile development with Kotlin and implementing frontend with
creative layouts and themes. Our CC members are able to build APIs and websites using
Google Cloud Platform. We estimated our project through 7 weeks ahead in this link:
https://bit.ly/C23-MP_Project-Timeline. We also attach our matchmaking procedure in this link :
https://bit.ly/MatchmakingProcedure.

You might also like