phddog-bot

import json
import re
from collections import Counter

# 1. Load corpus.json
with open("/home/ec2-user/serbia-discourse-graph/data/corpus.json", "r") as f:
    corpus = json.load(f)

# 2. Extract texts by discourse
discourse_names = {d["id"]: d["name"] for d in corpus["discourses"]}