twitter.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Analysing twitter archives"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "# Customise these variables for your own setup!\n",
    "TWEETS_PATH = \"archive_dir/twitter-archive-20180528/tweets.csv\"\n",
    "TWEETS_USERNAME = \"@caitelatte\"\n",
    "VERBOSE_RESULTS = True"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Setup: Read `tweets.csv`\n",
    "\n",
    "**Goal:** read the contents of the `tweets.csv` file into a list called `tweets`.\n",
    "\n",
    "### Notes\n",
    "\n",
    "-   I learnt that the `csv` module has a tool `DictReader` to read a CSV file as a dictionary, and pull out the field names from the first row. Each tweet is represented as an `OrderedDict`, which looks a lot like a list of tuples of field name and contents. This should be real nice to use!\n",
    "-   While working on optimising the first example I found a nice way to handle dates with the `matplotlib.dates` module."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "I found 4664 tweets!\n",
      "Each tweet will have the following fields: ['tweet_id', 'in_reply_to_status_id', 'in_reply_to_user_id', 'timestamp', 'source', 'text', 'retweeted_status_id', 'retweeted_status_user_id', 'retweeted_status_timestamp', 'expanded_urls']\n",
      "The first tweet in this list looks like:\n",
      "OrderedDict([('tweet_id', '1.00065E+18'),\n",
      "             ('in_reply_to_status_id', '1.00062E+18'),\n",
      "             ('in_reply_to_user_id', '6603532'),\n",
      "             ('timestamp', '2018-05-27 07:50:33 +0000'),\n",
      "             ('source',\n",
      "              '<a href=\"http://twitter.com/download/android\" '\n",
      "              'rel=\"nofollow\">Twitter for Android</a>'),\n",
      "             ('text',\n",
      "              '@b0rk My favourite tip is that you can replace the '\n",
      "              \"'/' with an arbitrary special character (I often use \"\n",
      "              \"#, there are a few you can't use) in sed commands so \"\n",
      "              \"you don't have to escape / in path names :)\"),\n",
      "             ('retweeted_status_id', ''),\n",
      "             ('retweeted_status_user_id', ''),\n",
      "             ('retweeted_status_timestamp', ''),\n",
      "             ('expanded_urls', '')])\n"
     ]
    }
   ],
   "source": [
    "import csv\n",
    "import pprint\n",
    "\n",
    "# load tweets from csv\n",
    "tweets = []\n",
    "with open(TWEETS_PATH) as tweetsfile:\n",
    "    tweetsReader = csv.DictReader(tweetsfile)\n",
    "    fields = tweetsReader.fieldnames\n",
    "    for tweet in tweetsReader:\n",
    "        tweets.append(tweet)\n",
    "\n",
    "if VERBOSE_RESULTS:\n",
    "    print(\"I found {0} tweets!\".format(len(tweets)))\n",
    "    print(\"Each tweet will have the following fields: {fields}\".format(\n",
    "        fields=fields\n",
    "    ))\n",
    "    print(\"The first tweet in this list looks like:\")\n",
    "    pprint.pprint(tweets[0], width=70)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "import matplotlib.dates as mdates\n",
    "import datetime\n",
    "\n",
    "def tweet_time_mdate(tweet_ts):\n",
    "    \"\"\"Return an mdate format of the timestamp in a tweet!\n",
    "    \n",
    "    Expects `tweet_ts` to be in format %Y-%m-%d %H:%M:%S %z.\n",
    "    \"\"\"\n",
    "    tweet_dt = datetime.datetime.strptime(tweet[\"timestamp\"], \"%Y-%m-%d %H:%M:%S %z\")\n",
    "    return mdates.date2num(tweet_dt)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Output 1: Basic tweet frequency\n",
    "\n",
    "**Goal:** a frequency histogram with the number of tweets per a given time period.\n",
    "\n",
    "### Notes:\n",
    "\n",
    "-   `datetime.datetime.strptime` is a function to parse a formatted string and return a `datetime.datetime` object. For documentation on how to format the format string, the best place to look is `help(time.strptime)`.\n",
    "-   I was originally going to write a list comprehension to enumerate each week between the first and last tweets available, but found out that `pandas` just *does this* with `pandas.date_range`. Amazing!\n",
    "-   After all that, it turns out I didn't need a date-range anyway, as matplotlib.pyplot actually has a `hist` function which auto-computes \"bins\" by itself.\n",
    "-   Dealing with all these tweets takes time if they're in `datetime.datetime` format... `matplotlib.dates` is a solution! I made a function above, `tweet_time_mdate` to take a string formatted like `2018-07-27 08:02:00 +0000` and return a `matplotlib.dates`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "I have collected 4664 datetimes into `tweet_times`, and the first one looks like (datetime.datetime(2018, 5, 27, 7, 50, 33, tzinfo=datetime.timezone.utc), datetime.date(2018, 5, 27), 736841.3267708333).\n"
     ]
    }
   ],
   "source": [
    "#  Now I want to generate a list of timestamps\n",
    "# (in python datetime.datetime format)\n",
    "import datetime\n",
    "import matplotlib.dates as mdates\n",
    "\n",
    "tweet_times = []\n",
    "tweet_dates = []\n",
    "tweet_mpldates = []\n",
    "for tweet in tweets:\n",
    "    tweet_ts = datetime.datetime.strptime(tweet[\"timestamp\"], \"%Y-%m-%d %H:%M:%S %z\")\n",
    "    tweet_times.append(tweet_ts)\n",
    "    tweet_dates.append(tweet_ts.date())\n",
    "    tweet_mpldates.append(tweet_time_mdate(tweet[\"timestamp\"]))\n",
    "\n",
    "if VERBOSE_RESULTS:\n",
    "    print(\"I have collected {0} datetimes into `tweet_times`, and the first one looks like {1}.\".format(\n",
    "        len(tweet_times), (tweet_times[0], tweet_dates[0], tweet_mpldates[0])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import datetime\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.dates as mdates\n",
    "\n",
    "\n",
    "# use this section for limiting numbers for performance\n",
    "tweet_mpldates_subset = tweet_mpldates[:500]\n",
    "\n",
    "# # Make a collection of frequency buckets with dates (not necessary due to plt.hist!)\n",
    "# import pandas.date_range\n",
    "# timediff = datetime.timedelta(days=7)\n",
    "# first_time = tweet_times[-1]\n",
    "# last_time = tweet_times[0]\n",
    "# print(first_time, last_time)\n",
    "\n",
    "# # Setting up a structure of date buckets for frequencies to go in\n",
    "# time_buckets = pandas.date_range(start=first_time, end=last_time, freq=timediff)\n",
    "# times_dict = []\n",
    "# for t_bucket in time_buckets:\n",
    "#     times_dict.append((t_bucket, 0))\n",
    "\n",
    "fig, ax = plt.subplots(1, 1)\n",
    "n, bins, patches = ax.hist(tweet_mpldates_subset, bins=24, stacked=True, color='teal')\n",
    "\n",
    "plt.xlabel(\"Dates\")\n",
    "plt.ylabel(\"Tweets\")\n",
    "plt.title(\"Histogram of Tweets by {username}\".format(\n",
    "    username=TWEETS_USERNAME))\n",
    "locator = mdates.AutoDateLocator()\n",
    "ax.xaxis.set_major_locator(locator)\n",
    "ax.xaxis.set_major_formatter(mdates.AutoDateFormatter(locator))\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Output 2: People interactions\n",
    "**Goal:** a representation of all the people I've mentioned over time.\n",
    "\n",
    "1. **Identify data structure goal:** list of tuples of usernames and timestamps\n",
    "2. **Extract the necessary data**:\n",
    "    - `USERNAME_REGEX = r\"(@\\w+)\"` (an @ symbol followed by word only characters)\n",
    "    - made the tweet_time_mdate function to avoid rewriting the same code as in the previous output\n",
    "    - tested with a slice of 100 tweets for speed\n",
    "    - `just_usernames` is a sorted list of all usernames\n",
    "    - `timed_usernames` is a list of tuples `(username, tweet_mdate)`\n",
    "3. **Present the data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "I've sorted 5305 username mentions!\n"
     ]
    }
   ],
   "source": [
    "import re\n",
    "\n",
    "# how can I recognise that a tweet has usernames?\n",
    "USERNAME_REGEX = r\"(@\\w+)\"\n",
    "\n",
    "timed_usernames = []\n",
    "just_usernames = []\n",
    "\n",
    "for tweet in tweets:\n",
    "    tweet_mdate = tweet_time_mdate(tweet.get(\"timestamp\"))\n",
    "    usernames = re.findall(USERNAME_REGEX, tweet.get(\"text\"))\n",
    "    for username in usernames:\n",
    "        timed_usernames.append((username, tweet_mdate))\n",
    "        just_usernames.append(username)\n",
    "        \n",
    "just_usernames = sorted(just_usernames)\n",
    "\n",
    "print(\"I've sorted {count} username mentions!\".format(\n",
    "    count=len(just_usernames)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# use this section for limiting numbers for performance\n",
    "usernames_subset = just_usernames[:100]\n",
    "uniq_usernames = set(usernames_subset)\n",
    "\n",
    "fig, ax = plt.subplots(1, 1)\n",
    "n, bins, patches = ax.hist(usernames_subset, bins=len(uniq_usernames), stacked=True, color='teal')\n",
    "\n",
    "plt.xlabel(\"Usernames\")\n",
    "plt.ylabel(\"Tweets\")\n",
    "plt.title(\"Histogram of tweets mentioning usernames by {username}\".format(\n",
    "    username=TWEETS_USERNAME))\n",
    "# locator = mdates.AutoDateLocator()\n",
    "# ax.xaxis.set_major_locator(locator)\n",
    "# ax.xaxis.set_major_formatter(mdates.AutoDateFormatter(locator))\n",
    "plt.xticks(usernames_subset, label=None, rotation='vertical')\n",
    "plt.axis(\"tight\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ok great! I made a graph, but this isn't very useful... Maybe a table would be better in this case?\n",
    "\n",
    "**New goal:** Display a table of the most frequent mentionees.\n",
    "\n",
    "**New tool:** numpy! specifically  [`numpy.unique`!](https://docs.scipy.org/doc/numpy/reference/generated/numpy.unique.html) - hinted from [Stack Overflow question \"Numpy - Most efficient frequency counts for unique values in an array?](https://stackoverflow.com/questions/10741346/numpy-most-efficient-frequency-counts-for-unique-values-in-an-array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The top 10 usernames I've mentioned are:\n",
      " @merxplat: 642\n",
      " @LibideeB: 617\n",
      " @caitelatte: 105\n",
      " @wjoneil: 88\n",
      " @ajdlinux: 85\n",
      " @scumsies: 68\n",
      " @helenperris: 66\n",
      " @clareology: 65\n",
      " @Ducky_Tape: 58\n",
      " @ANU_Events: 56\n",
      "There were 836 usernames I only mentioned once!\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "(uniq_usernames, count_usernames) = np.unique(just_usernames, return_counts=True)\n",
    "\n",
    "zipped_usernames_freq = []\n",
    "\n",
    "for zipzip in zip(uniq_usernames, count_usernames):\n",
    "#     print(zipzip)\n",
    "    zipped_usernames_freq.append(zipzip)\n",
    "\n",
    "# this sorts the counts as strings (eg '9' > '13')\n",
    "# as numpy makes their arrays homogeneous\n",
    "# print(np.sort(np.asarray((uniq_usernames, count_usernames))).T)\n",
    "# # also it's wrong???\n",
    "\n",
    "tuple_freqs_usernames = sorted(zipped_usernames_freq, key=lambda zipzip: zipzip[-1], reverse=True)\n",
    "\n",
    "print(\"The top 10 usernames I've mentioned are:\")\n",
    "for (user, count) in tuple_freqs_usernames[:10]:\n",
    "    print(\" {user}: {count}\".format(user=user, count=count))\n",
    "    \n",
    "print(\"There were {count} usernames I only mentioned once!\".format(\n",
    "    count=len(list(filter(lambda x: x[-1] ==1, zipped_usernames_freq)))\n",
    "))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"Make a histogram with just the top 100 usernames!\"\"\"\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# use this section for limiting numbers for performance\n",
    "usernames_subset, top_counts = zip(*tuple_freqs_usernames[:25])\n",
    "uniq_usernames = set(usernames_subset)\n",
    "\n",
    "fig, ax = plt.subplots(1, 1)\n",
    "n, bins, patches = ax.hist(usernames_subset, bins=len(uniq_usernames), weights=top_counts, stacked=True)\n",
    "\n",
    "plt.xlabel(\"Usernames\")\n",
    "plt.ylabel(\"Tweets\")\n",
    "plt.title(\"Histogram of tweets mentioning usernames by {username}\".format(\n",
    "    username=TWEETS_USERNAME))\n",
    "# locator = mdates.AutoDateLocator()\n",
    "# ax.xaxis.set_major_locator(locator)\n",
    "# ax.xaxis.set_major_formatter(mdates.AutoDateFormatter(locator))\n",
    "plt.xticks(usernames_subset, label=None, rotation='vertical')\n",
    "plt.axis(\"tight\")\n",
    "plt.set_cmap(\"ocean\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Output 3: Sentiment analysis!\n",
    "\n",
    "**TODO**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Notes/help messages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on built-in function sorted in module builtins:\n",
      "\n",
      "sorted(iterable, /, *, key=None, reverse=False)\n",
      "    Return a new list containing all items from the iterable in ascending order.\n",
      "    \n",
      "    A custom key function can be supplied to customize the sort order, and the\n",
      "    reverse flag can be set to request the result in descending order.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# all the help messages i needed\n",
    "# import datetime\n",
    "# help(datetime.datetime.strptime)\n",
    "# import time\n",
    "# help(time.strptime)\n",
    "# import matplotlib\n",
    "# help(matplotlib)\n",
    "# import datetime\n",
    "# help(datetime.datetime)\n",
    "help(sorted)"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}