Presented during the 2012 ACM Conference on Computer Supported Cooperative Work (CSCW 2012), the following original research on “the value of microblog content” is shared for your review and consideration.
Prepared by Paul Andre (Carnegie Mellon University), Michael Bernstein (MIT) and Kurt Luther (Georgia Institute of Technology), “Who Gives a Tweet: Evaluating Microblog Content Value” offers quantifiable insight into the perceived value of Twitter “tweets” through the lens of content, context and evolving social norms.
Primary questions considered as part of this milestone study of over 43,000 volunteer ratings on Twitter include:
“Conventional wisdom exists around these questions, but to our knowledge this is the first work to rigorously examine whether the commonly held truths are accurate. Further, by collecting many ratings, we are able to quantify effect sizes. A better understanding of content value will allow us to improve the overall experience of microblogging.” (Study Authors)
Predictors of “tweet” value in this study were based on “worth reading””neutral” or “not worth reading” ratings of individual tweets from eight specific categories that included.
Additionally, the reasons for determining whether a tweet was “liked” or “disliked” ranged from:
Reasons for Liking
Reasons for Disliking
PDF Version of Study: Click here.
Source: “Who Gives a Tweet: Evaluating Microblog Content Value” – Paul Andre (Carnegie Mellon University), Michael Bernstein (MIT) and Kurt Luther (Georgia Institute of Technology) – as prepared for CSCW’12, February 11–15, 2012, Seattle, Washington.
This entry was posted on Tuesday, February 21st, 2012 at 1:49 pm. It is filed under Blog Slider, chronology, discover, Live Feed and tagged with research, social media. You can follow any responses to this entry through the RSS 2.0 feed.
Comments are closed.
Taken from a combination of public market sizing estimations as shared in leading electronic discovery reports, publications and posts over time, the following eDiscovery Market Size Mashup shares general worldwide market sizing considerations for both the software and service areas of the electronic discovery market for the years between 2013 and 2018.
Provided as a non-comprehensive overview of key and publicly announced eDiscovery related mergers, acquisitions and investments to date in 2014, the following listing highlights key industry activities through the lens of announcement date, acquired company, acquiring or investing company and acquisition amount (if known).
The results presented here do not support the commonly advanced position that seed sets, or entire training sets, must be randomly selected [19, 28] [contra 11]. Our primary implementation of SPL, in which all training documents were randomly selected, yielded dramatically inferior results to our primary implementations of CAL and SAL, in which none of the training documents were randomly selected.
Cormack and Grossman set up an ingenious experiment to test the effectiveness of three machine learning protocols. It is ingenious for several reasons, not the least of which is that they created what they call an “evaluation toolkit” to perform the experiment. They have even made this same toolkit, this same software, freely available for use by any other qualified researchers. They invite other scientists to run the experiment for themselves. They invite open testing of their experiment. They invite vendors to do so too, but so far there have been no takers.
I want to talk about an issue that is attracting attention at the moment: how to select documents for training a predictive coding system. The catalyst for this current interest is “Evaluation of Machine Learning Protocols for Technology Assisted Review in Electronic Discovery”, recently presented at SIGIR by Gord Cormack and Maura Grossman.