History playground – finding patterns in historical newspapers

13 January 2020

share this:

Ever fancied finding out more about historical trends? Well, thanks to researchers at the University of Bristol, and their History Playground tool, anyone can analyse the content from a collection of historical British and American newspapers.

Macroscopic patterns of continuity and change over the course of centuries can be detected through the analysis of time series extracted from massive textual corpora. Similar data-driven approaches have already revolutionised the natural sciences. It is widely believed that there is similar potential for the humanities and social sciences. As such, new interactive tools are required to discover and extract macroscopic patterns from these vast quantities of data.

History Playground enables users to search for small sequences of words and retrieve their relative frequencies over the course of history. The tool makes use of scalable algorithms to first extract trends from textual corpora, before making them available for real-time search and discovery, presenting users with an interface to explore the data.

At present there are two large sets of text available:

Find out how to start using the History Playground by watching this short video:

Watch a further introduction to the project here:

History Playground uses the concept of n-grams, defined as short sequences of words. It is these n-grams that users search for when they use the tool. N-gram models are also widely used in the fields of natural language processing, probability, communication theory and data compression.

The team hope that in the long term, as more large textual datasets are released and additional feedback from the community helps to improve the Playground, they will be able to incorporate more varied and interesting corpora into the tool. In addition they are continuing to develop methods of analysis and additional views and visualisations. The tool also has the potential to incorporate text in languages other than English. For looking at more contemporary sources of data (for example, social media) the time resolution can be adjusted to study daily or even hourly changes.

This work is part of the ERC ThinkBIG project, Principal Investigator Nello Cristianini, University of Bristol.

Nello Cristianini is a Professor of Artificial Intelligence at the University of Bristol. His research interests include data science, artificial intelligence, machine learning, and applications to computational social sciences, digital humanities and news content analysis.



Read the full research articles on this topic:

Nello Cristianini is a Professor of Artificial Intelligence at the University of Bristol.

            AIhub is supported by:

Related posts :

Radical AI podcast: featuring Kate Crawford

In this episode Jess and Dylan chat to Kate Crawford about the Atlas of AI.
20 April 2021, by

AIhub – leading organisations form new charity to advance public understanding of artificial intelligence

AIhub aims to empower everyone to learn about AI. Learn more about our journey and new charity.
15 April 2021, by

Maximum entropy RL (provably) solves some robust RL problems

Nearly all real-world applications of reinforcement learning involve some degree of shift between the training environment and the testing environment, and it is increasingly important to learn policies that are robust to these changes.
14 April 2021, by

©2021 - Association for the Understanding of Artificial Intelligence