NLP project to read file and select specific table of document. Split the table row wise and create knowledge graph of it.
So the requirement is to build a project in python or jupyter notebook such that it reads text from pdf file(fitz library may suit the requirement).
Select the specific table with keywords matching in headers.(Can be achieved via headers filtering)
Lemmatize and pos tagging. Then build a knowledge graph sentence wise with noun as node and verb/adverb/adjective as edges.
The knowledge graph should be queryable once built maybe via py4neo or any other way. I do have an initial skeleton ipynb file with most of the reading file, extracting and cleanup.
3 days max as half of the work is done in the skeletal file I shall provide. 1 day would be awesome.
This is just a poc level for now. If the quality is good we can have an extended project soon with accuracy check and additional features soon.
Libraries n tools :
Recommend: fitz/pymupdf, spacy, nltk, Bert, networkx. Use any libraries that suit the project.
5 freelancers estão ofertando em média ₹9711 nesse trabalho
Hi, We have experienced Deep Learning expert who has experience in python programming, model training, processing structured and unstructured datasets. Please let me know if we can chat more on this. Thanks,