Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
7
datasets available to search
ShareScore release 0.9.0
Dataset results
7 results for “patent citations”
PatCit: A Comprehensive Dataset of Patent Citations
<p><em><strong>patCit: A Comprehensive Dataset of Patent Citations</strong></em> [<a href="https://tinyletter.com/patcit">Newsletter</a>, <a href="https://github.com/cverluise/PatCit">GitHub</a>]</p> <p>Patents are at the crossroads of many innovation nodes: science, industry, products, competition, etc. Such interactions can be identified through citations <em>in a broad sense</em>.</p> <p>It is now common to use front-page patent citations to study some aspects of the innovation system. However, <strong>there is much more buried in the Non Patent Literature (NPL) citations and in the patent text itself</strong>. <strong>patCit extracts and structures these citations.</strong></p> <blockquote> <p>Want to know more? Read patCit <a href="https://docs.google.com/presentation/d/11COlz64EZn8PipXvnDBBZI_bnDD0fpm6tyx1_EqD6lU/edit?usp=sharing">academic presentation</a> or dive into usage and technical guides on patCit <a href="https://cverluise.github.io/PatCit/">documentation website</a>.</p> </blockquote> <p><strong><em>IN PRACTICE</em></strong></p> <p>At patCit, we are building a <em>comprehensive</em> dataset of patent citations to help the community explore this <em>terra incognita</em>. patCit has the following features:</p> <ul> <li>global coverage</li> <li>front-page and in-text citations</li> <li>all categories of NPL documents</li> </ul> <p><strong><em>Front-page</em></strong></p> <p>patCit builds on <a href="https://www.epo.org/searching-for-patents/data/bulk-data-sets/docdb.html#tab-1">DOCDB</a>, the largest database of Non Patent Literature (NPL) citations. First, we deduplicate this corpus and organize it into 10 categories (bibliographical reference, database, norm & standard, etc). Then, we design and apply category specific information extraction models using <a href="https://github.com/explosion/spaCy">spaCy</a>. Eventually, when possible, we enrich the data using external domain specific high quality databases (e.g. Crossref for bibliographical references).</p> <p><strong><em>In-text</em></strong></p> <p>patCit builds on Google Patents corpus of <a href="https://console.cloud.google.com/bigquery?project=patcit-public-data&p=patents-public-data&d=patents&t=publications&page=table">USPTO full-text patents</a>. First, we extract patent and bibliographical reference citations. Then, we parse detected in-text citations into a series of category dependent attributes using <a href="https://github.com/kermitt2/grobid">grobid</a>. Patent citations are matched with a standard publication number using the Google Patents <a href="https://patents.google.com/api/match">matching API</a> and bibliographical references are matched with a DOI using <a href="https://github.com/kermitt2/biblio-glutton">biblio-glutton</a>. Eventually, when possible, we enrich the data using external domain specific high quality databases (e.g. Crossref for bibliographical references).</p> <p> </p> <p><strong>FAIR</strong></p> <p><strong>Find</strong> - The patCit dataset is available on <a href="https://console.cloud.google.com/bigquery?project=patcit-public-data&p=patcit-public-data&page=project">BigQuery</a> in an interactive environment. For those who have a smattering of SQL, this is the perfect place to explore the data. It can also be downloaded on <a href="https://zenodo.org/record/3710994">Zenodo</a>.</p> <p><strong>Interoperate</strong> - Interoperability is at the core of patCit ambition. We take care to extract unique identifiers whenever it is possible to enable data enrichment for domain specific high quality databases. This includes the DOI, PMID and PMCID for bibliographical references, the Technical Doc Number for standards, the Accession Number for Genetic databases, the publication number for PATSTAT and Claims, etc. See specific table for more details.</p> <p><strong>Reproduce</strong> - Our <a href="https://github.com/cverluise/PatCit">gitHub</a> repository is the project factory. You can learn more about data recipes and models on the patCit <a href="https://cverluise.github.io/PatCit/">documentation website</a>.</p>
Data and code for: Community structure in co-inventor networks affects time to first citation for patents
<p>This package provides the datasets and programming code needed to reproduce the results reported in the article "Community structure in co-inventor networks affects time to first citation for patents".</p> <p>v2: Added data and code pertaining to randomized-community-association test and updated README file.</p>
DATABASE: Electric Vehicle, Battery and Smart Grid patent citation networks and main paths.
<p>This dataset comprises the original patent citation networks that were created to calculate the main citation paths for the technologies of Electric Vehicle, Battery and Smart Grid.</p> <p>For each technology (1- Electric Vehicle, 2- Battery, 3- Smart Grid), four outputs are provided:</p> <p>a- Patent extraction: USPTO patents filtered by IPC or CPC and found in the Triadic Patent Families database (OECD, 2021) </p> <p>b- Full nodes and links reconstructed by following patent citations through a snowball method (until no further patents found)</p> <p>c- Filtered nodes and links according to keywords</p> <p>d- Main path nodes and links (with citation weights).</p> <p>For a detailed explanation of the methodology please refer to the submitted paper:</p> <p><strong>Transitions as a coevolutionary process: the urban emergence of electric vehicle inventions</strong></p>
Exploring the Impact of Negative Sampling on Patent Citation Recommendation
<ul> <li> <p><strong>pcr_patents.csv </strong>is the dataset which is generated by collecting samples randomly from Google Patents by exploiting a <a href="https://pypi.org/project/google-patent-scraper/">Python library</a>. The dataset comprises around 250,000 US patents and their titles, abstracts, and citations. Each patent has roughly on average 27 citations.</p> </li> </ul> <p>The zip file contains 3 different datasets for training and testing patent citation recommendation systems. These datasets were generated by utilizing the main dataset. They consist of around 1 million instances which are positive as well as negative samples. </p> <ul> <li> <p><strong>pcr_cpc_negative_sample_data.csv</strong> consists of negative samples that were generated based on CPC subclass codes. </p> </li> <li> <p><strong>pcr_random_negative_sample_data.csv</strong> consists of negative samples that were generated randomly. </p> </li> <li> <p><strong>pcr_sem_sim_negative_sample_data_2.csv</strong> consists of negative samples that were generated based on nearest neighbor relation.</p> </li> </ul>
Patent citation data for USPTO utility patents granted between 1976-2015 and for patents belonging to 30 technology domains
<p>These two data file contains information on patent citations for USPTO utility patents granted between 1976 and 2015 and for patents that have been classified in 30 specific technology domains.</p> <p>The file 'CITATION_INFO_no_neg_citlag.csv' is generated combining raw data <a href="https://www.patentsview.org/download/">freely dowloadable from patentsview.org</a> from which citations where the filing year of the citing patent is younger than the filing year of the cited one have been removed.</p> <p>The file 'CITATIONS_DOMAINS.csv' is a sample of the previous file that only includes citations made by patents belonging to one of 30 domains defined in the paper '<a href="https://www.sciencedirect.com/science/article/pii/S0040162520309264#ecom0001">Estimating technology performance improvement rates by mining patent data</a>' by Giorgio Triulzi, Jeff Alstott and Chris Magee.</p> <p>These two files complement another dataset <a href="http://dx.doi.org/10.17632/f4fj887y67.1">published on Mendeley Data</a>. The two datasets can be used, together with the code <a href="https://github.com/GiorgioTriulzi/TechnologyPerformanceImprovementEstimates">published on GitHub</a>, to replicate the main results from the paper.</p>
Research on key generic technology prediction based on graph neural networks under the perspective of patent citation - An example from the field of genetic engineering
<p>In this research, we adopted graph neural network models for key generic prediction based on cited patent data. Through the construction of the patent citation network and the design of a key generic evaluation system, 20879 relevant patents and 51,610 irrelevant patents were screened out. Further, we utilized the LDA topic model to interpret technical topics at a finer granularity. Finally, to test the effectiveness of this method, we took the field of genetic engineering as an example for key generic technology prediction, with an accuracy rate of 95%.</p>
Key generic technology prediction in patent citation using graph neural networks
Open the record for dataset details and reuse information.
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.