51社区黑料

Launched in 2020, SFU鈥檚 Social Data Analytics Minor Program (SDA) had the opportunity to host its first in-person event to promote the program, as part of the .

Taking place in a beautifully spacious room at 51社区黑料Burnaby鈥檚 Halpern building, close to 30 people sat in attendance to hear  speak on one of the many applications of data analysis to Economics, specifically with respect to nonmarket goods and climate change.

The SDA program is highly interdisciplinary, combining research fields like Linguistics, Political Science, Economics, and Philosophy. This interdisciplinarity was highlighted in Baylis鈥檚 talk, 鈥淎ssessing the impacts of environment hazards with text analysis鈥�.

As we continue to feel the effects of climate change, one topic economists like Baylis are interested in concerns the ways in which climate change affects nonmarket goods, goods like education, vaccination, or air quality, which cannot be bought and sold. Specifically, Baylis wanted to determine the ways in which climate change affects how people discuss extreme weather events and how that correlates with their levels of positive and negative emotions.

Although traditional methods exist for conducting such research, Baylis鈥檚 talk emphasized the limitations of those approaches, and鈥攊n line with the cutting-edge nature of the SDA program鈥攈ighlighted novel research methods to approach the issue.  

This novel research method involves collecting large amounts of Twitter data, and applying techniques from Natural Language Processing (NLP) and Statistics to gain valuable insights from the data. Specifically, Baylis鈥檚 work crucially relied upon conducting sentiment analysis on Twitter text-data to learn about how individuals react to changes in climate.

Sentiment analysis concerns the discernment of positive or negative opinions towards a topic through analyzing text, and happens to be a research speciality for 51社区黑料Linguistics , who is a member of the Steering Committee for the SDA program. In fact, the particular method of sentiment analysis used by Baylis is directly related to Dr. Taboada鈥檚  which further highlights the interdisciplinarity of the SDA program.

After Dr. Baylis鈥� talk, Dr. Taboada noted that it 鈥減rovided one of the best descriptions of the power of NLP in real-world applications.鈥�

Coming up:  Jennifer Hinnell, March 9

鈥淲hat鈥檚 in the hands? Analyzing speech and the body signal in spontaneous discourse鈥�

51社区黑料Linguistics students and faculty are encouraged to attend the next  on March 9th given by . Dr. Hinnell is a Killam postdoctoral research fellow at UBC鈥攁nd also a graduate from the MA Linguistics program here at SFU.  Her talk is titled 鈥淲hat鈥檚 in the hands? Analyzing speech and the body signal in spontaneous discourse鈥�.

In anticipation of her talk, we asked her to give us a preview. 鈥淲hat I鈥檓 going to talk about is how a multimodal approach to language answers important questions for the theoretical linguist, as well as moving forward methods from a data science perspective. I think one of the reasons that studying multimodal data has been largely excluded from linguistics is that it鈥檚 just really challenging. There鈥檚 so much variation. As in any study, the more variables you are trying to account for in the linguistic signal, the harder it is. But with tech advances over the last 5-10-20 years, we can do a lot more.鈥�

鈥淚鈥檓 so glad to be giving this talk at SFU, where I did my MA.  The research that I鈥檓 presenting next week is a natural progression from the work I did with Maite Taboada when we worked together on informal written language looking at internet blog data (back in 2010) and internet movie reviews across languages. Now I am able to add even more of the linguistic signal to the analysis of language-in-use, namely what the body is doing and how it contributes to communication when we are in spontaneous conversation.鈥�

Join us next week, Wednesday, March 9. 12:30 pm for the next SDA talk, with linguist, Jennifer Hinnell.  .