Please enable JavaScript.
Coggle requires JavaScript to display documents.
Challenges of Applying Machine Learning to Qualitative Coding (Grounded…
Challenges of Applying Machine Learning to Qualitative Coding
Grounded Theory
typical theory
theoretical framework
hyportheses
data
data
categories / organization
theory
(reverse way)
紮根理論編碼三層次
Open Coding
Axial Coding
Selective Coding
Qualitative Coding
open coding
no predefinted categories
closed (focus) coding)
tagging or mentioning
the candidate for support
the candidate for reject
specific hashtags
for support
for reject
Machine Learning for Qualitative coding
accuracy issue
Issue #1 Lack of mutual understanding leads to trust issues
Issue #2 Build a learning mode is not a primary goal
Issue #3 The nature conflict between qualitative and quantitative methods
Coding task
labor intensive
time consuming
Potential future research directions
1.Opening up the black box of ML
understandable
ML to qualitative coding
improve our understanding
interpretable
visualization
2.Reimaging the use of ML in qualitative
to adopt ML techiques
high accuracy
convincing social scientists
role
ML machines (students)
Humans (teachers)
not only in increasing accuracy ,but also in providing to reflect on the definitions of concepts