As discussed in the last meeting, we want to do an additional method of keyword extraction. In Issue #9 , the keyword extraction aims to get keywords directly related to the character. Additionally, we want to extract keywords from a larger context around the character in order to be able to describe in what settings they appear.
While step 3 remains the same as in #9, the finding of relevant text passages (1.) and the size of the context (2.) differs.
The problem of 1. is also addressed in #1 . However, it might also be sufficient to simply scan a chapter for the first (or second/third) mention of a name and then decide to extract keywords for this chapter.
For 2., I propose using the whole chapter in which the character appears.
As discussed in the last meeting, we want to do an additional method of keyword extraction. In Issue #9 , the keyword extraction aims to get keywords directly related to the character. Additionally, we want to extract keywords from a larger context around the character in order to be able to describe in what settings they appear.
While step 3 remains the same as in #9, the finding of relevant text passages (1.) and the size of the context (2.) differs.
The problem of 1. is also addressed in #1 . However, it might also be sufficient to simply scan a chapter for the first (or second/third) mention of a name and then decide to extract keywords for this chapter.
For 2., I propose using the whole chapter in which the character appears.