search

SvenElyes / Textanalytics

MIT License
1 stars 0 forks source link

Relations - Preprocessing - make text uniform #2

Open fmunzlin opened 3 years ago

fmunzlin commented 3 years ago

(1) Process methods which transfer the text like e.g. "Apple" --> "apple"; (2) Exchange e.g. "He" and "Her" with the last character in reference. (3) Deal with "die Jünger Jesu" = Andreas (er, wie sein Bruder Simon Petrus waren Jünger des Johannes), Simon (Kefas, Petrus), Jakobus (Sohn des Zebedäus), Johannes (Bruder des Jakobus - "Boanerges"), Philippus Bartolomäus (vielleicht der Jünger, der im Johannesevangelium Nathanael heißt) Simon (Kananäus), Matthäus (bei Markus auch "Levi" genannt), Thomas Jakobus (Sohn des Alphäus), Thaddäus - auch Judas, Sohn des Jakobs genannt), Judas (der Jesus verraten hatte) (4) Map between multiple character names, that are referencing the same character

SvenElyes commented 3 years ago

https://stackoverflow.com/questions/44714142/finding-relations-between-pronouns-and-nouns-in-sentences For (2)

aileen-reichelt commented 3 years ago

For (2), we need to do a co-reference resolution and for (3) and (4), we may do a synonym detection. We can probably use existing implementations for both