Meindl, Benjamin

Ott, Ingrid

Zierahn, Ulrich

2020-06-27T18:05:05Z

2019

<div class="csl-bib-body"> <div class="csl-entry">Meindl, B., Ott, I., &#38; Zierahn, U. (2019). Binary Patent Classification Methods for Few Annotated Samples. In <i>Proceedings of The 1st Workshop on Patent Text Mining and Semantic Technologies (PatentSemTech 2019)</i>. 1st Workshop on Patent Text Mining and Semantic Technologies (PatentSemTech 2019), Karlsruhe, Germany. https://doi.org/10.34726/pst2019.4</div> </div>

https://doi.org/10.34726/pst2019.4

http://hdl.handle.net/20.500.12708/831

In this paper, we develop binary patent classification algorithms for ambiguous concepts and small sample sizes. These are particularly useful for economic questions, which often require binary classification for implementing ambiguous and subjective concepts, where human classification is time-consuming, so that sample sizes are small. This covers examples such as whether workers are susceptible to automation or not, or whether a device is an automat or not. We compare the performance of naive Bayes, support vector machine, random forest and k-nearest neighbor classifiers with a the spaCy convolutional neural network (CNN) model, as well as spaCy CNN model pre-trained with patent data. The results show overall highest accuracy for the CNN models, with a significantly improved performance through pre-training. Our analysis suggests that the spaCy pre-trained CNN model provides a highly accurate NLP model, feasible for implementation without extensive computation capacity required. Pre-training was particularly beneficial for small sample sizes. Already 100 labeled patents lead to an accuracy of 77.2%. The low sample size required, may encourage researchers in various fields to use manually labeled patent data, for evaluating their specific question.

English

en

http://creativecommons.org/licenses/by/4.0/

patent classification

small sample size

convolutional neural network

language model pre-training

fast pre-training

Binary Patent Classification Methods for Few Annotated Samples

Inproceedings

Konferenzbeitrag

Creative Commons Namensnennung 4.0 International

Creative Commons Attribution 4.0 International

10.34726/pst2019.4

University of Lisbon, Portugal

Karlsruhe Institute of Technology, Germany

Centre for European Economic Research, Germany

Authors 2019

Full-Paper Contribution

Karlsruhe, Germany

Proceedings of The 1st Workshop on Patent Text Mining and Semantic Technologies (PatentSemTech 2019)

vor

AC15614257

5

urn:nbn:at:at-ubtuw:3-9430

CC BY 4.0

CC BY 4.0

1st Workshop on Patent Text Mining and Semantic Technologies (PatentSemTech 2019)

12-09-2019

12-09-2019

On Site

Event for scientific audience

Karlsruhe

DE

Meindl, Benjamin

en

conference paper

open

with Fulltext

Publications

http://purl.org/coar/resource_type/c_5794

Open Access

University of Lisbon

Karlsruhe Institute of Technology

Centre for European Economic Research