Become a sponsor to paco xander nathan
I'm developing open source libraries in Python for natural language and knowledge graph work, in particular:
- graph algorithms for lightweight unsupervised keyphrase extraction and entity linking
- knowledge graph representation and construction
- interoperability among discovery services for scholarly infrastructure
- metadata exchange about dataset usage and impact measures
- tutorials and curriculum materials about any of the above
Much of this focuses on the PyTextRank project and natural language courses, although other examples of my open source work include:
Previously I've worked in roles of open source community evangelism, conferences, etc., for Apache Spark and Project Jupyter. I continue to support these and other related projects, and mentor other people who are now in community evangelist or developer advocate roles.
Featured work
-
DerwenAI/pytextrank
Python implementation of TextRank algorithms ("textgraphs") for phrase extraction
Python 2,112 -
Coleridge-Initiative/rclc
Rich Context leaderboard competition, including the corpus and current SOTA for required tasks.
Python 21 -
DerwenAI/spaCy_tuTorial
A brief tutorial for spaCy 2.x, which runs on Google Colab
Jupyter Notebook 25 -
Coleridge-Initiative/RCApi
Rich Context API integrations for federating metadata discovery and exchange across multiple scholarly infrastructure providers
Python 11 -
DerwenAI/disparity_filter
Implements a disparity filter in Python, based on graphs in NetworkX, to extract the multiscale backbone of a complex weighted network (Serrano, et al., 2009)
Python 29 -
ceteri/synecdoche
A collection of random things which are overtly textual in nature.
Python 2