keep track of timeline collection times and size for pseudonym/list

[berli0z]

We want to extract and visualize data that helps monitoring so called "bot" facebook profiles as the ones used in the italian elections to see when are we capturing data and if everything is running smoothly.

• In order to do this we will write a python script that will use APIs and given a user pseudonym will retrieve anonymous information about its activity (impressions collected and at what times).
• Allow to keep track of different bots given a list of user pseudonyms
• Add hierarchical configuration
• visualize this via the telegram bot and maybe through a web intreface,
• produce a document that explains how to use it. Details about this should be included in the texts related to setting up bots for facebook impressions collection in eu19.

Some code:

import matplotlib.pyplot as plt
from datetime import datetime
import matplotlib.dates as mdates
import numpy as np

date_format = "%Y-%m-%dT%H:%M:%S.%fZ"

#CSV file downloaded from "Your Data" page
file = 'file.csv'

df = pd.read_csv(file)

timelinelist = []
for index, row in df.iterrows():
    id = row['timeline']
    if not id in timelinelist:
        timelinelist.append(id)
        print(id)
    else:
        pass

maxlist = []
dateslist = []
for timeline in timelinelist:
    df2 = df[df['timeline'] == timeline]
    maxim = df2.impressionOrder.values.max()
    maxlist.append(maxim)
    minim = df2.impressionTime.values.min()
    dateslist.append(datetime.strptime(minim, date_format))
    del df2

results = list(zip(dateslist, maxlist))

data = pd.DataFrame(results, columns= ['time', 'impressions'])
today = datetime.now()
data = data[data['time'] > today.date()]
plt.plot(data.time, data.impressions)
plt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%H:%M'))
plt.gca().xaxis.set_major_locator(mdates.AutoDateLocator())
plt.yticks(np.arange(0, data.impressions.values.max()+10, step=5))
plt.show()```