Infrastructure and Communities Minister Catherine McKenna promoted the SAMbot on Twitter Wednesday to detect and track toxic sentiment, primarily towards women.

During the 2021 Canadian federal election, SAM, or Samara Areto Monitorbot, will collect data and insights about the online abuse directed at party leaders and incumbent candidates running for re-election across the country.

The Samara Centre for Democracy website reads, "While it is commonly understood that toxic online spaces are harming our democracy, there is little data that illustrates the extent of the problem in detail. Consequently, effective regulations, policies and social expectations for online conduct in Canadian political contexts are lacking."

SAM is a machine learning bot that collects and analyzes tweets in real-time and provides a tweet analysis while storing each tweet that mentions the Twitter handle of at least one of the candidates being monitored.

Online political discourse during campaign periods can be "extremely toxic," and the 2021 federal election is a "unique opportunity" for SAM to investigate the current state of Canada’s online political space.

The website adds that the information SAM provides can "inform important conversations" and "nuanced approaches" to reducing the toxicity of online political spaces in Canada.

SAM will track all English and French tweets directed at political party leaders and incumbent candidates. SAM tracks messages — whether a reply, quote, tweet or mention — will be analyzed on seven-point toxicity attributes.

The organization claims that Twitter data is collected and used according to Twitter’s acceptable terms of use.

SAM analyzes and stores tweets by analyzing the tweet information to identify and extract the tweet text, scoring the tweet on a seven-point toxicity scale; and, storing the tweet text and its toxicity score in a database.

In a separate tweet, McKenna said, "The misogyny women journalists and politicians face on Twitter is nauseating, exhausting and completely unacceptable," referring to a tweet by Global News Mercedes Stephenson on why women leave public life.

"The misogyny on Twitter is literally nauseating to me. If you’re going to attack a woman in public, you better check your facts first," she wrote. "I have no conflict of interest, and if I did, I would declare it. I can’t believe I even have to tweet this."

Stephenson added that this is why women leave public life, citing "the insane social media abuse."

She has been targeted in a series of social media attacks "at an ever-increasing rate" since she broke the sexual misconduct in the military story. "The irony is not lost on me that people call themselves progressive while doing this," Stephenson concluded.

McKenna also wrote, "And if you're one of the haters, get lost."

A policy brief titled Women, politics and Twitter: Using machine learning to change the discourse explained the technology used and its rationale. "Our political systems are unequal, and we suffer for it," wrote the brief.

"Twitter is an important social media platform for politicians to share their visions and engage with their constituents," but adds, "Women are disproportionately harassed on this platform because of their gender."

To raise awareness of online abuse and shift the discourse surrounding women in politics, the researchers designed, built, and deployed ParityBOT: a Twitter bot that classifies hateful tweets directed at women in politics and then posts "positivitweets."

The researchers deployed ParityBOT during the 2019 Alberta provincial election, and the 2019 Canadian federal election. For each tweet we collected, we calculated the probability that the tweet was hateful or abusive."

During the 2019 Alberta provincial election, the model classified 1468 tweets of the total 12726 as hateful and posted only 973 positivitweets.