Compassion.ly Machine Learning

Focusing on the machine learning aspects of our project Compassion.ly, which is an app that will help people find their major and university.

Tasks

• Data Collection:

  • Collect data from the LTMPT website and gather as much information about majors as possible.
  • Find details about each major, including the courses they offer and the career prospects for specific majors.

• Data Labeling:

  • Label the dataset and add weights to the data for each major according to specific topics:
    • Matematika
    • Sains
    • Fisika
    • Sosiologi
    • Biologi
    • Kimia
    • Teknologi
    • Bisnis dan Ekonomi
    • Seni
    • Sastra dan Linguistik
    • Pendidikan
    • Hukum
    • Lingkungan
    • Kesehatan
    • Geografi
    • Komunikasi
    • Sejarah dan Filsafat

• Modeling:

  • Build three models:
    • Recommendation Model: Use content-based filtering with TensorFlow and Scikit-learn cosine similarity. access in this link content-based-filtering-model
    • Similarity Search: Fine-tune the T5 architecture to find similar texts for the majors. The input for this model is text, allowing the model to predict the best matches and answers. t5 similarity search
    • Text Summarization: Fine-tune the T5 architecture for text summarization. text_summarization

Model architecture

• Content-based filtering using Tensorflow and cosine similarity

  

• t5 architecture for similarity search and text summarization

  

Dataset

For the dataset, access on this link: dataset_compassionly

Model demo

The demo for our model, you can access in this link: Model demo

Deployed Model

For our deployed model, you can access it in this link:

• Summarization Model
• Quick Recommendation Model
• Numerical Model as an Input

Model evaluation

This is our model accuracy and loss

Reference :

• R.Manjula (2016). Content Based Filtering Techniques in Recommendation System using user preferences. International Journal of Innovations in Engineering and Technology (IJIET)
• Pradeep, et al (2020). A Machine Learning approach for automation of Resume Recommendation system, Procedia Computer Science
• Parul, et al (2017). Comparing Content Based and Collaborative Filtering in Recommender Systems. International Journal of New Technology and Research (IJNTR).
• Sri Hari Nallamala, et al (2020). A Brief Analysis of Collaborative and Content Based Filtering Algorithms used in Recommender Systems. IOP Conference Series: Materials Science and Engineering.
• Shijie Geng, et al (2022). Recommendation as Language Processing (RLP): A Unified Pretrain, Personalized Prompt & Predict Paradigm (P5).
• Jiacheng Li, et al (2023). Text Is All You Need: Learning Language Representations for Sequential Recommendation. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD ’23).
• Radhwan Hussein (2023). Empowering Education: AMCQA Generation with T5 Transformers. ISAR - International Journal of Mathematics and Computing Techniques.