The Radical Right and the Threat of Violence
Jonathon Morgan (September 2016)
Paper’s reference in the IEEE style?
J. Morgan, “The Radical Right and the Threat of Violence,” The Radical Right and the Threat of Violence, 08-Sep-2016. .
How did you find the paper?
Referenced from Partially Derivative podcast1)http://partiallyderivative.com/
If applicable, write a list of the search terms you used.
Was the paper peer reviewed? Explain how you found out.
This is a blog post
Does the author(s) work in a university or a government-funded research institute? If so, which university or research institute? If not, where do they work?
The author is a data scientist, records a podcast on data science claled Partially Derivative and is the Founder and CEO of New Knowledge, a data science startup
What does this tell you about their expertise? Are they an expert in the topic area?
Potentially experienced in NLP but not a theoretical expert.
What was the paper about?
The paper summarises analysis completed into the rise and radicalization of 'alt-right' (e.g. anti-semitic, anti-gay, white supremecists) commentary in social media as a result of Donald Trump's presidential campaign.
Twitter profile descriptions and images were analysed to identify 'alt-right' users and posts. Posts from a group of 3,500 users were then analysed and quantified in terms of racism, xenophobia, anti-Semitic or violent tendencies etc.
This data was then analysed to find similarities between normal compared to 'alt-right' use of words. As an example in normal posts the word 'Jewish' is most similar to other religion words like 'Christian', 'Muslim', 'Arabs', etc but in the 'alt-right' posts 'Jewish' was most similar to words like 'Homosexual', 'Anti-white', Satanic', Homosexual' etc, more likely indicating a sentiment of hate.
A radicalisation score was determined and plotted over time. This indicated a significant increase in the level of radicalisation since the beginning of 2016.
If applicable, is this paper similar to other papers you have read for this assignment? If so, which papers and why?
Different from previous papers
If applicable, is this paper different to other papers you have read for this assignment? If so, which papers and why?
Different from previous papers
What do these similarities and differences suggest? What are your observations? Do you have any new ideas? Do you have any conclusions?
This is the first in-depth use of NLP to analyse social media language i've reviewed.
The analysis demonstrates an interesting example of what can be achieve with social media.
I followed up with the author to better understand his analysis. In summary, he scraped twitter for the 'alt-right' data, trained a model using Word2Vec using it, and compared similarity between the model and the Google News corpus model.
target_group = target_word2vec_model.most_
similar('Jewish') regular_usage = ground_truth_word2vec.most_ similar('Jewish') ground_truth_word2vec.cosine_ distance(target_group, regular_usage)
If I have time, I may try to replicate this with commentary about legalising gay marriage and the plebicite in Australia.
This question is to be answered after your critical analysis is completed. Which sections (if any) of your critical analysis was this paper cited in?
References [ + ]