In poldham/opsrdev: Access Patent Data with the European Patent Office "Open Patent Services" (OPS) API

knitr::opts_chunk$set(echo = TRUE)

Get started with RStudio

RStudio works with projects.

In the top right of the windown you will see a blue sign with R. Click on that.

1. New Project
2. Give the Project a Name (you will need this later and it will be public)
3. Select a location to save the project (such as desktop or my documents). Then OK.
4. Go to Build in the top menu
5. Choose Configure Build Tools . Choose website.

Ok... we are good to go.

opsrdev session

We need to load some libraries. You can find each one in Packages and tick them or you can write the following in the Console.

library(opsrdev) # the opsr package
library(dplyr)
library(readr) 

Authenticate

OPS will work for very limited searches without authenticating.

When you signed up you were asked to obtain a key and a secret. You can save them in R to use later by typing the following in the console.

Note that the sign is <- and the key and secret must be inside quotes "".

key <- "BiaLwkHwuus8WjXQPdvVajGGrLMkZnf9"
secret <- "wkvTOQeuLwUGYxcw"

Next we can use this information to authenticate

library(opsrdev)
ops_auth(key, secret)

What is the status of the service (it can be busy)

library(opsrdev)
ops_status()

How many results are there

We can only retrieve upto 2000 results at a time.

library(opsrdev)
ops_count("drones")

Ok that's fine. If we had larger results we would want to break them down by year. Let's try pizza.

library(opsrdev)
ops_count("pizza")

library(opsrdev)
ops_count("pizza", "ta", start = 1990, end = 2015)

Still too many. Let's try a narrower set.

library(opsrdev)
ops_count("pizza", "ta", start = 2010, end = 2015)

So, by experimenting we can arrive at a date range under the 2000 results that we could then retrieve.

APIs work with Urls

APIs by definition work with URls.

The basic url for the EPO OPS service is

http://ops.epo.org/3.1/rest-services

We won't use this but just to show you how this works programatically. The ops_urls function will create the urls for you.

urls <- ops_urls(query = "pizza", type = "ti", start = 1990, end = 2000)
urls

We want to fetch back some bibliographic data. We can use a function called ops_fetch_biblio(). As the name suggests it fetches bibliographic data. It uses a set of controls where type = "ta" is titles and abstracts, while type = "ti"

library(opsrdev)
drones <- ops_fetch_biblio(query = "drones", type = "ta", service = "biblio", start = 1990, end = 2015, timer = 10)

Let's see that as a table in RStudio.

View(drones)

We can also write this data to a .csv file. It will write the file to your project folder.

write_csv(drones, "drones.csv") 

Don't forget quotes or .csv at the end.

Congratulations, you have now accessed data from an API using R and downloaded the results.

poldham/opsrdev documentation built on May 25, 2019, 11:23 a.m.