Stanford NLP Model Output for Biofuel Patent Classification (doi:10.7910/DVN/29374)

View:

Part 1: Document Description
Part 2: Study Description
Part 5: Other Study-Related Materials
Entire Codebook

Stanford NLP Model Output for Biofuel Patent Classification

doi:10.7910/DVN/29374

Harvard Dataverse

2015-03-06

1

Kessler, Jeff, 2015, "Stanford NLP Model Output for Biofuel Patent Classification", https://doi.org/10.7910/DVN/29374, Harvard Dataverse, V1

Stanford NLP Model Output for Biofuel Patent Classification

doi:10.7910/DVN/29374

Kessler, Jeff (University of California, Davis)

Harvard Dataverse

Harvard Dataverse Network

Jeff Kessler

2015-03-06

2015

https://doi.org/10.7910/DVN/29374

Biofuel Classifier

Natural Language Processing

This NLP model was generated using the Stanford NLP Classifier (available from: http://nlp.stanford.edu/software/classifier.shtml). The model was trained using a random selection of 700 manually classified biofuel patents from 1976 through 2013, and validated against 300 manually classified biofuel patents on January 03, 2014. Included are the classification results and associated patent numbers for both the manually trained patents, and for the automatically categorized patents.

1976-2013

United States

<a href="http://creativecommons.org/publicdomain/zero/1.0">CC0 1.0</a>

Label:

Manual Classification.csv

Text:

This is the initial list of 1000 patents manually classified for use with training and validating the NLP model

text/plain; charset=US-ASCII

Label:

ner-model.ser.gz

Text:

This is the model generated by the Stanford NLP Classifier

application/x-gzip

Label:

NLP Classification.csv

Text:

This is the list of patents and associated classifications based on the NLP model that was trained using the manually classified patents

text/plain; charset=US-ASCII

Label:

patents_test.prop

Text:

This is the property file used for parameterizing the model

text/plain; charset=US-ASCII