London: British linguistics experts will analyse over 2.3 billion words, compiled from speeches made by lawmakers over the past 200 years, to determine the change parliamentary language has undergone on issues like honesty, war and terrorism.
The study by researchers at the University of Glasgow hopes to understand when and why certain topics are raised in parliament and how their use and significance evolved over time.
The experts will also try to study how the language of individual MPs changed during their tenure in office.
The researchers will use data from 1803 to 2003 to chart the popularity of various topics like honesty, honour, homosexuality, war and terrorism.
The 15-month project will make use of Hansard, the official Westminster records which contains transcripts of all parliamentary debates in Britain.
The project is split into two phases. The researchers will first develop a computer programme capable of sorting through and aggregating the huge amounts of lexical data.
Then, the team will use the software to hone in on key phrases and concepts that recur over time.
By measuring the frequency and context of the appearances, they will build a detailed analysis of the shifting concerns of parliament over the past two centuries.
Marc Alexander, lead investigator from the University of Glasgow, said: "What this project is allowing us to do is develop a more sophisticated analysis technique for searching through big sets of data and identifying what`s important. Looking at Hansard will hopefully reveal big-scale patterns in our society and culture.
"We will use it to analyse how parliament anticipates or reacts to major global events and social issues. We can even use it to look at how the language of individuals, such as Thatcher or Churchill, changed over the course of their political careers."
The technology developed as part of the SAMUELS (Semantic Annotation and Mark-Up for Enhancing Lexical Searches) project can also be applied to other texts, opening up the potential for future research on many different texts.
The project is being funded by the Arts and Humanities Research Council (AHRC) and Economic and Social Research Council (ESRC) in Big Data.