LucidAi / nlcd

News Life Cycle Detector
MIT License
3 stars 1 forks source link

Implement explicit reference detection #9

Open zaycev opened 10 years ago

zaycev commented 10 years ago

One idea is to define explicit document reference as follows:

If document D1 contains substring S (sentence or quote) from document D2 and source or author of D2 is mentioned in D1 within a window of N sentences to the left and to the right of S, then D1 explicitly refers to D2. We can start with N=3.

For example:

D2 = {text='This is a great article!', source="BBC",author=} D1 = {text='BBC wrote: "This is a great article!"'}