Feature #1815
Updated by Nandini Bansal about 3 years ago
In continuation of #1810
We need to implement a new context matching algorithm for the KPs matching with <word1.word2> subsection headers. We will make use of parent header information of the KPs and similar docs to match/check the context. The intuition is that KP doc and similar doc falling under the same group/hierarchy of the headers have a higher probability of being related to each other as compared to other similar docs.
Save the parent header info in the text file as well to identify patterns in the KP and similar doc.