Enhancing our understanding of Montréal’s communities through the use of computer-assisted text analysis
Looking back on the fifth Montréal in Common webinar
Written by Amélie Ducharme and Julien Vallières,* research assistant and coordinator for the Laboratory for Collective Narratives and Discourse Analysis
On May 25th, 2022, the Laboratory for Collective Narratives and Discourse Analysis (LACONDA) team participated in the fifth Montréal in Common webinar, during which they presented their work within the Data for Society Hub. The primary purpose of this presentation was to share an update on the developing narrative monitoring bulletin, which uses computer-assisted text analysis as a means to efficiently process an ever-growing body of documentary sources.
By processing natural language, assisted text analysis aims to organize and diminish the amount of documentation, and to facilitate the interplay of automated and manual methods, thereby shaping the analyst’s work by means of a processing chain. It enables the automation of several tasks related to text analysis, assists the analyst in carrying out their work, and helps develop a monitoring bulletin which uses automatic language processing to facilitate information monitoring.
LACONDA researchers are currently developing a proof-of-concept for these methods based on the media coverage of food insecurity in Montréal. They hope to demonstrate how the use of these computerized methods can benefit Montréal’s communities. The team has already assembled an extensive text corpus, and is now focusing on the indexing and geo-referencing of these materials. They will also begin to analyze the resulting social representations of food insecurity. By 2024, the researchers expect to have the ability to formalize the analysis processing chain, to measure representation gaps, and to have made headway in the automation of the process.
While there remain legal and institutional barriers to overcome, project director Pascal Brissette believes it is necessary to provide the city’s civic actors with a decision-making tool that integrates the advances of automatic language processing to better understand the communities they are serving.
Watch a replay of the webinar below:
* Content editing: Elisabeth Doyon, Pascal Brissette, Karolyne Arseneault and Julie Levasseur.
The Data for Society Hub is a project by Montréal in Common, a community for the development of innovative projects for the Smart Cities Challenge.