Topic Modelling (LDA) and Text Generation (GPT-3) with Environmental/Sustainable Development Goals (SDG) Data

This repo covers Topic Modelling and Text Generation. For the Topic Modelling task, the LDA model was used, while a replication of GPT-3 was used for Text Generation. Of course a replication of it was used due to limited resources. The task focuses on environmental issues: protecting forests, biodiversity and more

Useful Links

1. You can find the OSDG Community dataset on this site: https://zenodo.org/record/7136826#.Y5s8nX1BxPa

2. You can find the medium article on this through this link: https://medium.com/@watipasomulwafu/topic-modelling-lda-and-text-generation-gpt-3-on-sustainable-development-goals-sdg-data-ab3e6b0300bc

3. You can also access the code through this google colab link: https://colab.research.google.com/drive/1zJl4h70JZ4lQjOiD1PA3cAXpvql7Syz3?usp=sharing

Using Google Coloab?

1. Download the dataset: https://zenodo.org/record/7136826#.Y5s8nX1BxPa
2. Upload it to your Google drive root folder
3. Mount it as seen in the first cell of the code

Let's get started! 😉