algorithm - Determine the difficulty of an english word -
i working word based game. word database contains around 10,000 english words (sorted alphabetically). planning have 5 difficulty levels in game. level 1 shows easiest words , level 5 shows difficult words, relatively speaking.
i need divide 10,000 long words list 5 levels, starting easiest words difficult ones. looking program me.
can tell me if there algorithm or method quantitatively measure difficulty of english word?
i have thoughts revolving around using "word length" , "word frequency" factors, , come formula or accomplishes this.
get large corpus of texts (e.g. gutenberg archives), straight frequency analysis, , eyeball results. if don't satisfying, weight each text flesch-kincaid score , run analysis again - words show frequently, in "difficult" texts score boost, want.
if have 10000 words, though, quicker frequency sorting first pass , tweak results hand.
Comments
Post a Comment