Report Postclose
Remove Postclose
Are you sure? After you remove the post, it will no longer appear in channel listings but you can access it directly. You can undo this later by clicking "approve".
Delete Postclose
Are you sure you want to delete this post? This is a permanent action and cannot be reversed.
I have a goal. I am in the initial phase of determining if my goal is feasible. I would like to ask MIT Open readers for advice on books, and maybe one has already tried to accomplish what I desire.
In about 1 year, I plan to attend and complete Caltech and IBM's 7 months bootcamp on AI/ML. In project 1, we will learn how to use Natural Language Processing (NLP) to determine online social networks (OSNs) posts for hate speech. I would like to extend this idea to climate action.
Specifically, I would like to mine OSNs for text regarding the psychology and financial impacts of climate disasters. From research, I would suspect psychology to change from grief to anxiety to worry as one moves away from the moving epicenter of the climate disaster. I expect financial impact to be worse for the poor and people of color according to literature.
I have read that corpus linguistics (CL), discourse analysis (DA), and critical discourse analysis (CDA) can be used to understand power and inequality relationships. I am curious if I can use the latter, especially CDA, along with NLP, to derive a psychological power (severity) and sentiment analysis of OSN posts related to climate disasters. Then the data can be used to predict future psychological and inequality scenarios in post climate disaster locations from another moving epicenter.
Will I be able to do this for various locations in the state of Florida as an example? Can I use this AI model to quickly predict where psychological and financial help will be needed post disaster?
If I can accomplish my goal, can I use this data to impact citizens desire to actively support climate action? As an example, a scenario where an equivalent hurricane would hit another area of Florida, affect the victims according to the AI prediction, and show, through AI, what the people should expect. Would this drive them to contact their politicians in advance and demand climate action? Would the inequality issue and power relationships cause the news media to hold politicians responsible before an event even occurred?
I am just learning. So, I have a LOT more questions than the above. I am learning the strengths and weaknesses of each method I have discussed, but it will take me time. Do you have any book suggestions, literature suggestions, or practical experience? I have nearly 1.58 years before I try to apply my idea, and [1-10] are some of my references so far.
References:
[1] Price, H. (2022). The Language of Mental Illness: Corpus Linguistics and the Construction of Mental Illness in the Press. United Kingdom: Cambridge University Press.
[2] Brindle, A. (2016). The Language of Hate: A Corpus Linguistic Analysis of White Supremacist Language. United Kingdom: Routledge, Taylor & Francis Group.
[3] Gwen Bouvier & David Machin (2018) Critical Discourse Analysis and the challenges and opportunities of social media, Review of Communication, 18:3, 178-192, DOI: 10.1080/15358593.2018.1479881
[4] Agarwal N. Dokoohaki N. & Tokdemir S. (2019). Emerging research challenges and opportunities in computational social network analysis and mining. Springer.
[5] Neuman, Y. (2022). How to Find a Needle in a Haystack. Taylor & Francis. https://bookshelf.vitalsource.com/books/9781000787115
[6] Garrick, Jacqueline; Buck, Martina. (September 2020). Whistleblower Retaliation Checklist: A New Instrument for Identifying Retaliatory Tactics and Their Psychological Impacts After an Employee Discloses Workplace Wrongdoing. Crisis, Stress, and Human Resilience. Vol. 2, No. 2.
[7] van der Velden, P. G., Pecoraro, M., Houwerzijl, M. S., & van der Meulen, E. (2019). Mental Health Problems Among Whistleblowers: A Comparative Study. Psychological Reports, 122(2), 632–644. https://lnkd.in/gXvKqhXa
[8] Hardie, A., McEnery, T. (2011). Corpus Linguistics: Method, Theory and Practice. (n.p.): Cambridge University Press.
[9] Tamássy, R., & Géring, Z. (2022). Rich variety of DA approaches applied in social media research: A systematic scoping review. Discourse & Communication, 16(1), 93–109. https://doi.org/10.1177/17504813211043722
[10] Lachlan O'Neill, Nandini Anantharama, Wray Buntine and Simon D Angus (2021), SoDa Laboratories Working Paper Series No.
2021-12, Monash Business School, available at http://soda-wps.s3-website-ap-southeast-
2.amazonaws.com/RePEc/ajr/sodwps/2021-12.pdf
I am looking at social responses to the environmental crisis and I have found ecolingusitics (a type of critical discourse analysis) to use as my methodology. There is also something called ecocriticism. I didn't know how to look at social media - too much? and am looking at texts across various disciplines - to identify certain responses (environmental positions = from ecolingusitics, and ecosophies = personal philosophies towards the environment). If you could use AI to select and analyse social media posts it would help a lot.
Thanks. I worry that part of the problem with online social network analysis is that, as an example, tweets are often grammatically incorrect because of constraints on text length. I hope to learn if the latter is or is not an issue when I take the Caltech and IBM course.
I just read from [A] in the last hours. Linguistics is so fascinating. As Professor Price explains, corpus linguistic analysis removes a lot of the biases that, as an example, Likert scales will create. It seems to be very powerful.
Part of my desire is to mine online social networks to develop keywords that are more representative of the mind and attitude of the population. Questionnaires and Likert analysis seem to be less accurate when developing keywords. I don't know if AI can do this currently. I know corpus linguistics software can.
References:
[A] Price, H. (2022). The Language of Mental Illness: Corpus Linguistics and the Construction of Mental Illness in the Press. United Kingdom: Cambridge University Press. [edited by author]
Hi. I have found from my reading about ecolinguistics that they use certain 'positions' as indications of specific ecosophies, and in ecocriticism images are used. I am not sure how that is going to translate into keywords. I am looking more at phrases perhaps. I found Aron Stibbe wrote a lot about Ecolingistics and Greg Garrard wrote a book on Ecocriticism. Maybe their work can help you a bit. I would like to see how you manage to extract useful information from social media, because I couldn't figure out how to do it ):
I can't help you right now, but I will ask your question about collecting phrases when I attend the Caltech and IBM bootcamp.
Also, this is a book that I plan to read in the future[A]. I will remember you as I read. I just ask that you phrase your question here so that I can record it.
References:
[A] Data Mining Approaches for Big Data and Sentiment Analysis in Social Media. (2021). United States: IGI Global.
I would caution against assumptions such as "I expect financial impact to be worse for the poor and people of color according to literature." That may be the perception some want and thus may be the 'think', but with the just passed IRA the opposite is closer to the truth. This is one of the struggles with Climate Action -- the oil barons who want to keep their oil profits are working hard to convince everyone that Climate Action will cost huge amounts of $$ where in fact it will save $$. For example, with the IRA many Climate Action steps in Buildings won't even cost more up front because of large tax credits that are even refundable for non-profits ... then those upgrades effectively save/make $$ forever! The key is that all those windfall profits that the oil industry has and continues to steal from the rest of us will disappear only to become our investments in Renewable Energy and Energy Efficiency.
Thanks for your advice!