Knowledge is everywhere on the internet but they not well structured for users to digest faster.
What it does
Learnit is a knowledge graph powered app that helps learn faster and easier about Islam. For every entity, one can see all semantic relationships connected to the entity, for the use case I used data about the prophets in Islam. For every prophet, one can see information like Place of birth, Place of Death, Spouse, Children. This approach helps learn faster without having to go through 1000words articles before I could learn about a prophet or maybe a scholar I am researching about.
How I built it
- Data was gotten from Wikipedia using SPARQL as the query language.
- The data was then modeled into graph schematically using Cypher.
- The design of the app was made in Figma.
Challenges I ran into
Data is not readily available, I had to contribute to Wikipedia open source in order to get the correct label for the SPARQL query.
Accomplishments that I'm proud of
- I was able to contribute to Wikipedia data during data ingestion.
- I was also able to explore more with cloud infrastructures like GCP and Azure while trying to set up my neo4j database on the cloud.
What I learned
I learned how to use FIGMA for a simple design and also learned more about wiki RDF
What's next for LEARNit
- Integrate autocomplete search
- Run NLP named entity extraction model on articles
- Ingest more data for production-ready kind of app