GitHub

Finding placenames This project illustrates an application to find new placenames in a dataset, by performing NER on Geograph image descriptions and comparing those with an Ordnance Survey gazetteer. It uses a simple local index to compare gazetteer entries, and only allows exact matching. It is available in Binder for experimentation:

Comparing documents This project illustrates an application to compare file content using cosine similarity. A simple version build binary vectors of parts of speech, and calculates cosine similarity on that basis, while a second version uses Spacy's inbuilt word2vec version. You can try it out here:

Name		Name	Last commit message	Last commit date
Latest commit History 65 Commits
.ipynb_checkpoints		.ipynb_checkpoints
data		data
Basic_NLP.ipynb		Basic_NLP.ipynb
Gazetteer.ipynb		Gazetteer.ipynb
LICENSE		LICENSE
Postings.ipynb		Postings.ipynb
README.md		README.md
SpatialIndex.ipynb		SpatialIndex.ipynb
SpatialSearchDemo.ipynb		SpatialSearchDemo.ipynb
Untitled.ipynb		Untitled.ipynb
Untitled1.ipynb		Untitled1.ipynb
Untitled2.ipynb		Untitled2.ipynb
geo871_gazetteer.ipynb		geo871_gazetteer.ipynb
geo871_placename_experiments.ipynb		geo871_placename_experiments.ipynb
geo871_similarity.ipynb		geo871_similarity.ipynb
geo871_similarity_wiki.ipynb		geo871_similarity_wiki.ipynb
geograph_corpus-Copy1.ipynb		geograph_corpus-Copy1.ipynb
geograph_corpus.ipynb		geograph_corpus.ipynb
picture_comparisons.ipynb		picture_comparisons.ipynb
requirements.txt		requirements.txt
runtime.txt		runtime.txt
tf_idf.ipynb		tf_idf.ipynb

License

rspurves/geo871Experiments

Folders and files

Latest commit

History

Repository files navigation

About

Resources

License

Stars

Watchers

Forks

Languages