Isabel Holmes
A
In response to Adam Deitsch □.
Excited to listen! I had a go asking it some social psychology questions and it sounded great but got a lot wrong. I worry that students will think it’s given them a great essay and then fail the course.
02:48 PM - Feb 22, 2023
0
1
Isabel Holmes
A
In response to Clare Healy.
Late to the party but hi from Northamptonshire!
06:05 AM - Feb 06, 2023
1
1
Isabel Holmes
A
In response to Spencer M. Ross.
Thanks! We actually had interesting results re readability. Based on anecdotal evidence (which I’ll admit fitted my own narrative 😬) we thought dems would score higher. But actually the reps wrote at a higher level! Would be interesting to know if that’s just a Reddit thing or not
02:27 AM - Feb 06, 2023
0
1
Isabel Holmes
A
In response to Isabel Holmes.
Not sure if threads are a thing here so I won’t do the usual academic unnecessarily long breakdown (find that on the hellscape if you want it). But feedback on dealing with nerves re some glaring error I accidentally might have made appreciated 😬
05:05 PM - Feb 05, 2023
0
0
Isabel Holmes
A
In response to Isabel Holmes.
We looked at whether taking an interdisciplinary approach (social psych) to feature selection could be helpful for model accuracy when classifying redditors by political affiliation. We also tested our model for generalisability across topic
05:03 PM - Feb 05, 2023
1
1
loading...
{{ notificationModalContent }} {{ promptModalMessage }}