[Air-L] NLP for sentiment analysis of social media comment

[Caroline Haythornthwaite]

Some references that will help on this topic.

Chapter 32: Sentiment Analysis, Mike Thelwall, In The SAGE Handbook of Social Media Research Methods. DOI: http://dx.doi.org/10.4135/9781473983847.n32
— and other chapters in this book. 

My review of the book briefly summarizes what’s in each chapter — I recommend the whole collection!
Haythornthwaite, C. (2017). Review of Luke Sloan and Anabel Quan-Haase (Eds.)(2017) “The SAGE Handbook of Social Media Research Methods”, Thousand Oaks, CA: SAGE. Journal of the Association of Information Science and Technology, 69(9), 625-630.. http://onlinelibrary.wiley.com/doi/10.1002/asi.23957/full.

Other work
See that of Anatoliy Gruzd and his social media lab (http://socialmedialab.ca/author/gruzd/), and the Netlytic program for text analysis (http://socialmedialab.ca/apps/netlytic/)

With Anatoliy and colleagues we have just finished a paper on the human coding side of NLP (but not gone on to the automated analysis yet) of some Reddit subreddits. This is the forthcoming paper (also a HICSS conf paper from Jan 2018, but the paper has results of applying the coding). 

Kumar, P., Gruzd, A., Haythornthwaite, C., Gilbert, S., Esteve del Valle, M. & Paulin, D. (January 2018). Learning in the wild: Coding Reddit for learning and practice.  51st Hawaii International Conference on System Sciences, Big Island, HI. 

Haythornthwaite, C., Kumar, P., Gruzd, A., Gilbert, S., Esteve Del Valle, M., & Paulin, D. (2018, in press). Learning in the Wild: Coding for Learning and Practice on Reddit. Learning, Media and Technology. https://doi.org/10.1080/17439884.2018.1498356

/Caroline

> Date: Sat, 14 Jul 2018 14:51:08 +0000
> From: Nina Lasek <Nina.Lasek1 at hotmail.com>
> To: "air-l at listserv.aoir.org" <air-l at listserv.aoir.org>
> Subject: [Air-L] NLP for sentiment analysis of social media comment
> Message-ID:
> 	<VI1PR0202MB35045F32EFFEBDFF29E8D915A85F0 at VI1PR0202MB3504.eurprd02.prod.outlook.com>
> 	
> Content-Type: text/plain; charset="iso-8859-1"
> 
> Dear all,
> 
> most sentiment analyses of social media comments I know use dictionary-based approaches (e.g. sentigstrength). However, I am wondering if researchers also use Natural Language Processing approaches to examine the sentiment of social media comments; I am still an absolute beginner in this field, hence I would be very glad if somebody could point me to useful papers/tutorials/software for doing NLP based sentiment analyses?
> 
> Many thanks,
> Nina
> 
> 
> ------------------------------
> 
> Message: 3
> Date: Sat, 14 Jul 2018 15:15:21 +0000
> From: "Astvansh, Vivek" <astvansh at iu.edu>
> To: Nina Lasek <Nina.Lasek1 at hotmail.com>, "air-l at listserv.aoir.org"
> 	<air-l at listserv.aoir.org>
> Subject: Re: [Air-L] NLP for sentiment analysis of social media
> 	comment
> Message-ID: <1531581325827.65343 at iu.edu>
> Content-Type: text/plain; charset="iso-8859-1"
> 
> Hi, Nina:
> 
> Yes; some scholars (particularly in the past, when field-specific dictionaries, such as one in finance, were not available) used their own supervised machine-learning method. This method comprised: (a) using human beings (e.g., MTurkers) to classify text into positive, negative, and neutral sentiment, (b) using this human classification to train a classifier (Lasso or support vector machine), and lastly, (c) using the trained model to classify the holdout sample. As you can guess, this is quite some work and perhaps not required when the relevant dictionary is readily available and you have a program that uses this dictionary to classify text.
> 
> You can search Google Scholar to find articles that use human annotations and machine-learning classifiers and follow this supervised machine-learning approach.
> 
> Best wishes!
> Vivek Astvansh
> Assistant Professor of Marketing
> Kelley School of Business, Indiana University Bloomington
>