recentpopularlog in

driscoll : analysis   91

« earlier  
Women as Background Decoration: Part 1 - Tropes vs Women in Video Games - YouTube
In our latest episode we use examples from 52 different games released between 1986 and 2014.
Gaming  representation  sexism  gender  women  analysis  feminism 
june 2014 by driscoll
How Netflix Reverse Engineered Hollywood - Alexis C. Madrigal - The Atlantic
ICYMI this week @alexismadrigal's story on reverse engineering Netflix genres, cherry-topped with my genre generator
netflix  data  mining  text  film  recommendation  algorithm  bigdata  cluster  analysis 
january 2014 by driscoll
What Would I Say?
"Yup, this wasn't relevant although rather magical" - auto-gen status updates from past status updates
text  analysis  bot  socialmedia  privacy  mining  datamining  data  bigdata  art  netart 
november 2013 by driscoll
A Method of Automated Nonparametric Content Analysis for Social Science | Gary King
A Method of Automated Nonparametric Content Analysis for Social Science

Citation:
Hopkins, Daniel, and Gary King. "A Method of Automated Nonparametric Content Analysis for Social Science." American Journal of Political Science 54, no. 1 (2010): 229-247. copy at http://j.mp/jNFDgI

Export
Abstract:

The increasing availability of digitized text presents enormous opportunities for social scientists. Yet hand coding many blogs, speeches, government records, newspapers, or other sources of unstructured text is infeasible. Although computer scientists have methods for automated content analysis, most are optimized to classify individual documents, whereas social scientists instead want generalizations about the population of documents, such as the proportion in a given category. Unfortunately, even a method with a high percent of individual documents correctly classified can be hugely biased when estimating category proportions. By directly optimizing for this social science goal, we develop a method that gives approximately unbiased estimates of category proportions even when the optimal classifier performs poorly. We illustrate with diverse data sets, including the daily expressed opinions of thousands of people about the U.S. presidency. We also make available software that implements our methods and large corpora of text for further analysis.
twitter  bigdata  data  nlp  methodology  method  text  analysis  analytics  academia  research 
september 2012 by driscoll
« earlier      
per page:    204080120160

Copy this bookmark:





to read