LSE - Small Logo
LSE - Small Logo

Sabrina Argoub

April 5th, 2021

A new tool by Google puts the power of AI in the hands of reporters

0 comments | 6 shares

Estimated reading time: 5 minutes

Sabrina Argoub

April 5th, 2021

A new tool by Google puts the power of AI in the hands of reporters

0 comments | 6 shares

Estimated reading time: 5 minutes

Analysing vast datasets of written or audio content in search of a story can be a daunting task for reporters. Pinpoint, a research tool from Google Journalist Studio promises to simplify the process. Megan H. Chan – News Ecosystem Lead at the Google News Initiative – and Matt Kiefer – Data Specialist at The Washington Post – explain how Pinpoint is helping reporters to find more stories, faster. 

Key Takeaways

  • Pinpoint enables reporters to upload large document sets, make them public if they wish, and analyse them at pace. A wide variety of document formats are compatible, including email archives, audio files, images, and hand-written documents.  
  • Entity extraction finds names, places, institutions in large document sets and automatically connect similar searching terms alongside one another. For example, Pinpoint can recognise “JFK”, “John F Kennedy”, and “John F Kennedy Junior” as the same search term. 
  • Zipcodes, synonyms, and text embedded within images can be included in search instructions, according to the needs of the reporter. 
  • Audio files, once uploaded, are turned into PDF transcripts. Search terms are then highlighted and references made to time codes in the audio files.


“We built Pinpoint to put AI into the hands of all types of journalists and to empower newsrooms without big technical teams.”


Megan H. Chan, News Ecosystem Lead, Google News Initiative

Pinpoint is part of Journalist Studio, a suite of Google tools designed to make the work of journalists simpler and more efficient. Leveraging the latest machine learning techniques, these tools make swift work of analysing large document sets and help journalists make their work more secure and creative.

Megan H. Chan – News Ecosystem Lead, Google News Initiative

Pinpoint has been built to give an answer to the question of “How can we make the work of journalists more efficient?”. Going through huge document sets, routinely sourced and received by reporters, is an arduous and meticulous task. 

Pinpoint uses Google Knowledge Graph and extracts the most recognisable entities. It also includes a function that enables documents to be shared between journalists and collaborate with colleagues. 

In partnership with a number of organisations like The Washington Post and Document Cloud, Google has curated a number of collections that are already on the platform for journalists to use. In addition, journalists can upload up to 200k documents – of varying file type and size – to their own workspace.

 

 

How Journalists Use Pinpoint

Some examples of journalists putting Pinpoint to good use include Maria Ressa’s team at Rappler, who used Pinpoint to upload 13k recently released CIA documents. They were able to process them in 30 minutes, then write and publish a story about Manila in record time. 

Hearst Media Group used Pinpoint to go through a dataset of cases of child sexual abuse. They uploaded hundreds of documents from hundreds of court cases and were able to find common terms, places, and names. 

Reveal at Centre for Investigative Reporting uploaded an internal email archive from ICE to analyse COVID testing in these facilities and compare it with official accounts.

The Boston Globe used the synonym search capabilities to investigate what type of equipment the police are recording as buying, where terms are often unknown to journalists, and common and less common terms are used interchangeably.  

Verificado in Mexico compared the transcripts of the daily press briefings from the Mexican President about COVID with their own recordings, which enabled them to perform fact-checking at pace. They uploaded their own archives to Pinpoint for cataloguing. 

 

 

Matt Kiefer – Data Specialist, The Washington Post

The Washington Post was among the first ones to test Pinpoint. Matt Kiefer – part of a team intent on pushing the boundaries of journalism and storytelling through experimentation – outlines that Pinpoint is a useful lever for technical teams but also the broader newsroom at large. 

Pinpoint lets journalists maintain a really important tenet of the investigative practice: showing the work behind the story. By sharing document sets with the wider public and the journalist community, others can get involved and search through documents, finding new angles and stories that corroborate the findings. 

The Washington Post collects documents from a number of events and sources and uploads them to Pinpoint. For example, the findings of the Mueller investigation, the Afghanistan Papers and Secret Service receipts for Trump invoiced family trips, and much more.

Additional resources

 


 

This article was written by Georgia Meyer, POLIS intern and LSE’s MSc student, and edited by Sabrina Argoub, JournalismAI Community Coordinator.

The article includes the main takeaways from one of the sessions of the 2020 JournalismAI Festival. You can watch all the Festival recordings on YouTube.

JournalismAI is a project of POLIS and it’s powered by the Google News Initiative.

About the author

Sabrina Argoub

Posted In: Expert Insights | JournalismAI

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.