The backend API connecting with Cosmos DB (Gremlin), to power the graph database visulization of linked data.
The customizable scripts can also ingest free texts and semi-structued data into a graph network.
(pseudonymized linked data of oncology study from different sources)
From multidimensional relational databases and freetext reports to a semantic network of clinics, diagnosis (pathological, radiological, genomics), deseases (tumors, metastasis), treatments (surgery, radiotherapy, medication), responses and assessments data.
(a glimps of the middle east around 2700 years ago)
Wikipedia articles are recording events for each calendar year. The intention is to generate a knowledge graph of historical events of the entier human history... or only for the parts properly recorded in Wikipedia :-)
The HTML pages are relatively structured starting from 700 BC, For example https://en.wikipedia.org/wiki/700_BC.
A web scrapper is built to extract data from these pages, and store information as nodes/edges in the graph database.