Some english words get parsed as epithets which never exist as real epithets, e.g. the. Add a blacklist to the parser that avoids such words ever become epithets.
Not as easy as one thinks as the main parsing is one big regex and its match depends on submatches. So if "the" should not match, the regex needs to know that. Otherwise other parts like authors are wrongly matched too
mdoering
commented
5 years ago
Some english words get parsed as epithets which never exist as real epithets, e.g.
the. Add a blacklist to the parser that avoids such words ever become epithets.Not as easy as one thinks as the main parsing is one big regex and its match depends on submatches. So if "the" should not match, the regex needs to know that. Otherwise other parts like authors are wrongly matched too