DMCA
Teaching the Computer to Code Frames in News: Comparing Two Supervised Machine Learning Approaches to Frame Analysis
| Citations: | 1 - 0 self |
BibTeX
@MISC{Burscher_teachingthe,
author = {Björn Burscher and Daan Odijk and Rens Vliegenthart and Maarten De Rijke and Claes H. De Vreese},
title = {Teaching the Computer to Code Frames in News: Comparing Two Supervised Machine Learning Approaches to Frame Analysis},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We explore the application of supervised machine learning (SML) to frame coding. By automating the coding of frames in news, SML facilitates the incorporation of large-scale content analysis into framing research, even if financial resources are scarce. This furthers a more integrated investigation of framing processes conceptually as well as methodologically. We conduct several experiments in which we automate the coding of four generic frames that are operationalised as a set of indicator questions. In doing so, we compare two approaches to modelling the coherence between indicator questions and frames as an SML task. The results of our experiments show that SML is well suited to automate frame coding but that coding performance is dependent on the way SML is implemented. In most framing studies, news frames are coded with indicator questions in manual Content Analysis (CA) (Matthes, 2009). Generally, measures of several indicators are combined to cover different aspects of a frame (e.g., Simon & Xenos, 2000). Human coders can be properly trained to code frame indicators, and through training their performance can be improved until accu-racy and reliability reach satisfactory levels. However, human coding is a time-consuming and costly process. This limits the scope of CA in framing research. Computers, in contrast, are more naturally suited for the processing of large quantities of documents and the repetitiveness of cod-
Keyphrases
supervised machine learning approach code frame indicator question frame analysis several indicator costly process supervised machine learning manual content analysis sml task satisfactory level several experiment large-scale content analysis financial resource generic frame human coding way sml integrated investigation simon xenos different aspect frame indicator framing study large quantity human coder news frame
