analysis.py

import csv, datetime, statistics


def timestamp(s):
    return int(datetime.datetime.strptime(s, '%Y-%m-%d %H:%M:%S').timestamp())


with open('QueryResults.csv') as f:
    lst = list(csv.DictReader(f))

tags = {}

for record in lst:
    tag_name = record["TagName"]
    parent_id = record["ParentId"]
    question_creation = timestamp(record["QuestionCreation"])
    answer_creation = timestamp(record["AnswerCreation"])
    time_passed = answer_creation - question_creation
    
    if tag_name in tags:
        if parent_id in tags[tag_name]:
            if time_passed < tags[tag_name][parent_id]:
                tags[tag_name][parent_id] = time_passed
        else:
            tags[tag_name][parent_id] = time_passed
    else:
        tags[tag_name] = {parent_id: time_passed}

means = {}

for tag, dct in tags.items():
    means[tag] = int(statistics.mean(dct.values()))

for key, val in sorted(means.items(), key=lambda r: r[1]):
    print('   ', key, ' ' * (50 - len(key)), val)