Extract decisions from documents

Hi all,

My name is Nick du Chatinier and I recently found out about Neo4J during a meeting with the company Fluid. Since they are working a lot with Neo4J and NLP I am very interested and formulated some of my own ideas.

Unfortunately, I got stuck on this idea. The idea can actually be described as follows. In the Netherlands, all government meetings are transcribed. These transcripts also contain the decisions about motions submitted and whether they have been adopted or not. I would like to investigate how I can read these documents and extract the decisions (if any). I would like to see which party has voted for and which against.

On the basis of this information, the next step would be to investigate whether the parties are actually sticking to their election plan at all.

When I look for the use of NLP to analyze this, I don't really come across anything concrete. Apparently it is possible to find the most common words with TextRank?

I will give a small example of the text in which a motion was passed:

The Segers / Klaver motion (35570, no. 26) was put to the vote.

Chairman:
I note that the members of the political groups of the SP, the PvdA, Krol, GroenLinks, the PvdD, 50PLUS, Van Kooten-Arissen, DENK, D66, the VVD, the SGP, the CDA, the ChristenUnie, the PVV and Van Haga have voted in favor of this motion and the members of the FvD faction against it, so that it is adopted. "

Can someone help me and point me in the right direction? I am eager to learn but simply cannot find what I need to use to do this correctly.