Given a document in English, how can you estimate the ease with which someone will find they can read it? Does it require a college-level of reading comprehension or is it something a much younger student could read and understand?
While these questions are useful to ask, they don't admit a simple answer. One option is to use one of the (essentially identical) two Flesch Kincaid Readability Tests. These are simple calculations which provide you with a rough estimate of the reading ease.
In this episode, Kyle shares his thoughts on this tool and when it could be appropriate to use as part of your feature engineering pipeline towards a machine learning objective.
For empirical validation of these metrics, the plot below compares English language Wikipedia pages with "Simple English" Wikipedia pages. The analysis Kyle describes in this episode yields the intuitively pleasing histogram below. It summarizes the distribution of Flesch reading ease scores for 1000 pages examined from both Wikipedias.
A Long Way Till AGI
Brain Inspired AI
Computable AGI
AGI Can Be Safe
AI Fails on Theory of Mind Tasks
AI for Mathematics Education
Evaluating Jokes with LLMs
Why Machines Will Never Rule the World
A Psychopathological Approach to Safety in AGI
The NLP Community Metasurvey
Skeptical Survey Interpretation
The Gallup Poll
Inclusive Study Group Formation at Scale
The PhilPapers Survey
Non-Response Bias
Measuring Trust in Robots with Likert Scales
CAREER Prediction
The Panel Study of Income Dynamics
Survey Design Working Session
Bot Detection and Dyadic Surveys
Create your
podcast in
minutes
It is Free
Insight Story: Tech Trends Unpacked
Zero-Shot
Fast Forward by Tomorrow Unlocked: Tech past, tech future
The Unbelivable Truth - Series 1 - 26 including specials and pilot
Acquired