Reddit Sentiment.ipynb

reddit-sentiment.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Reddit Sentiment.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyMOevKSlcC0kH7pjBSwnNvv",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/firmai/6a89400120fb9480c687ed719cc44a98/reddit-sentiment.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Starting (API/Raw Data)"
      ],
      "metadata": {
        "id": "d8YLsVeGw7NE"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "At the start you have either raw data or a type of API that you have access to. Our purpose here is to create a numerical sentiment database from textual data and other features."
      ],
      "metadata": {
        "id": "0DUB2gmRw_nA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install praw"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5bBa9QWkya5Z",
        "outputId": "f2c385e9-237e-49e8-b7d1-b9c417ef051b"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting praw\n",
            "  Downloading praw-7.7.1-py3-none-any.whl (191 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m191.0/191.0 kB\u001b[0m \u001b[31m4.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting prawcore<3,>=2.1 (from praw)\n",
            "  Downloading prawcore-2.4.0-py3-none-any.whl (17 kB)\n",
            "Collecting update-checker>=0.18 (from praw)\n",
            "  Downloading update_checker-0.18.0-py3-none-any.whl (7.0 kB)\n",
            "Requirement already satisfied: websocket-client>=0.54.0 in /usr/local/lib/python3.10/dist-packages (from praw) (1.7.0)\n",
            "Requirement already satisfied: requests<3.0,>=2.6.0 in /usr/local/lib/python3.10/dist-packages (from prawcore<3,>=2.1->praw) (2.31.0)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests<3.0,>=2.6.0->prawcore<3,>=2.1->praw) (3.3.2)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests<3.0,>=2.6.0->prawcore<3,>=2.1->praw) (3.6)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests<3.0,>=2.6.0->prawcore<3,>=2.1->praw) (2.0.7)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests<3.0,>=2.6.0->prawcore<3,>=2.1->praw) (2024.2.2)\n",
            "Installing collected packages: update-checker, prawcore, praw\n",
            "Successfully installed praw-7.7.1 prawcore-2.4.0 update-checker-0.18.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "VADER ( Valence Aware Dictionary for Sentiment Reasoning) is a model used for text sentiment analysis that is sensitive to both polarity (positive/negative) and intensity (strength) of emotion."
      ],
      "metadata": {
        "id": "9d7M7A_5JIx1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "## This could take 5-10 minutes to load\n",
        "import nltk\n",
        "from nltk.sentiment.vader import SentimentIntensityAnalyzer as SIA\n",
        "import praw\n",
        "import matplotlib.pyplot as plt\n",
        "import math\n",
        "import datetime as dt\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "\n",
        "\n",
        "nltk.download('vader_lexicon')\n",
        "nltk.download('stopwords')\n",
        "\n"
      ],
      "metadata": {
        "id": "soTxrx2ZFSyM",
        "outputId": "60833bbc-f20d-4362-c525-b887623795ca",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[nltk_data] Downloading package vader_lexicon to /root/nltk_data...\n",
            "[nltk_data]   Package vader_lexicon is already up-to-date!\n",
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Package stopwords is already up-to-date!\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {},
          "execution_count": 1
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "6UWDPIBvtJHJ"
      },
      "outputs": [],
      "source": [
        "\n",
        "reddit = praw.Reddit(client_id='gtvkIEzRvVUf-U65JaQlNg',\n",
        "                    client_secret='6mGollB6YVWWdNk56CLELFSoi8diMA',\n",
        "                    user_agent='mathieudempsey',\n",
        "                     check_for_async=False) ## to use this, make a Reddit app. Client ID is in top left corner, client secret is given, and user agent is the username that the app is under\n",
        "\n",
        "\n",
        "\n",
        "sub_reddits = reddit.subreddit('wallstreetbets')\n",
        "stocks = [\"GME\", \"PTON\"]\n",
        "\n",
        "def commentSentiment(ticker, urlT):\n",
        "    subComments = []\n",
        "    bodyComment = []\n",
        "    try:\n",
        "        check = reddit.submission(url=urlT)\n",
        "        subComments = check.comments\n",
        "    except:\n",
        "        return 0\n",
        "\n",
        "    for comment in subComments:\n",
        "        try:\n",
        "            bodyComment.append(comment.body)\n",
        "        except:\n",
        "            return 0\n",
        "\n",
        "    sia = SIA()  # VADER’s SentimentIntensityAnalyzer() takes in a string and returns a dictionary of scores (negative, neutral, positive)\n",
        "                 # .. and compound which is computed by normalising the three before-mentioned scores. (positive is high, negative low)\n",
        "    results = []\n",
        "    for line in bodyComment:\n",
        "        scores = sia.polarity_scores(line)\n",
        "        scores['headline'] = line\n",
        "\n",
        "        results.append(scores)\n",
        "\n",
        "    df =pd.DataFrame.from_records(results)\n",
        "    df.head()\n",
        "    df['label'] = 0\n",
        "\n",
        "    try:\n",
        "        df.loc[df['compound'] > 0.1, 'label'] = 1\n",
        "        df.loc[df['compound'] < -0.1, 'label'] = -1\n",
        "    except:\n",
        "        return 0\n",
        "\n",
        "    averageScore = 0\n",
        "    position = 0\n",
        "    while position < len(df.label)-1:\n",
        "        averageScore = averageScore + df.label[position]\n",
        "        position += 1\n",
        "    averageScore = averageScore/len(df.label)\n",
        "\n",
        "    return(averageScore)\n",
        "\n",
        "\n",
        "def latestComment(ticker, urlT):\n",
        "    subComments = []\n",
        "    updateDates = []\n",
        "    try:\n",
        "        check = reddit.submission(url=urlT)\n",
        "        subComments = check.comments\n",
        "    except:\n",
        "        return 0\n",
        "\n",
        "    for comment in subComments:\n",
        "        try:\n",
        "            updateDates.append(comment.created_utc)\n",
        "        except:\n",
        "            return 0\n",
        "\n",
        "    updateDates.sort()\n",
        "    return(updateDates[-1])\n",
        "\n",
        "\n",
        "def get_date(date):\n",
        "    return dt.datetime.fromtimestamp(date)\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "This could be very slow, and even take 10 mins or so to run."
      ],
      "metadata": {
        "id": "_b1ook01aW61"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "submission_statistics = []\n",
        "d = {}\n",
        "for ticker in stocks:\n",
        "    for submission in reddit.subreddit('wallstreetbets').search(ticker, limit=130): #Search submissions related to the ticker\n",
        "        if submission.domain != \"self.wallstreetbets\":\n",
        "            continue\n",
        "        d = {}\n",
        "        d['ticker'] = ticker\n",
        "        d['num_comments'] = submission.num_comments\n",
        "        d['comment_sentiment_average'] = commentSentiment(ticker, submission.url)\n",
        "        if d['comment_sentiment_average'] == 0.000000:\n",
        "            continue\n",
        "        d['latest_comment_date'] = latestComment(ticker, submission.url)\n",
        "        d['score'] = submission.score\n",
        "        d['upvote_ratio'] = submission.upvote_ratio\n",
        "        d['date'] = submission.created_utc\n",
        "        d['domain'] = submission.domain\n",
        "        d['num_crossposts'] = submission.num_crossposts\n",
        "        d['author'] = submission.author\n",
        "        submission_statistics.append(d)\n",
        "\n",
        "dfSentimentStocks = pd.DataFrame(submission_statistics)\n",
        "\n",
        "_timestampcreated = dfSentimentStocks[\"date\"].apply(get_date)\n",
        "dfSentimentStocks = dfSentimentStocks.assign(timestamp = _timestampcreated)\n",
        "\n",
        "_timestampcomment = dfSentimentStocks[\"latest_comment_date\"].apply(get_date)\n",
        "dfSentimentStocks = dfSentimentStocks.assign(commentdate = _timestampcomment)\n",
        "\n",
        "dfSentimentStocks.sort_values(\"latest_comment_date\", axis = 0, ascending = True,inplace = True, na_position ='last')\n",
        "\n",
        "\n",
        "dfSentimentStocks.to_csv('Reddit_Sentiment_Equity.csv', index=False)"
      ],
      "metadata": {
        "id": "9cch2LaNN-R8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "try:\n",
        "  dfSentimentStocks = pd.read_csv(\"Reddit_Sentiment_Equity.csv\", parse_dates=['commentdate'])\n",
        "except:\n",
        "  dfSentimentStocks = pd.read_csv(\"https://storage.googleapis.com/public-quant/course//content/Reddit_Sentiment_Equity(1).csv\", parse_dates=['commentdate'])"
      ],
      "metadata": {
        "id": "RCTAf7phZRfE"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "del dfSentimentStocks[\"latest_comment_date\"]\n",
        "del dfSentimentStocks[\"date\"]"
      ],
      "metadata": {
        "id": "rEFl-NuiMWqE"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dfSentimentStocks = dfSentimentStocks.set_index(\"commentdate\")"
      ],
      "metadata": {
        "id": "OTT8okKRYFjc"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dfSentimentStocks[\"ticker\"].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Sn2QrqSCFfmO",
        "outputId": "f63c026e-a4d6-4fd7-9272-05b09330eea9"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "PTON    51\n",
              "GME     31\n",
              "Name: ticker, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "dfSentimentStocks = dfSentimentStocks[dfSentimentStocks[\"ticker\"]==\"PTON\"]"
      ],
      "metadata": {
        "id": "stzonsLTYyJ8"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dfSentimentStocks = dfSentimentStocks.resample('W')"
      ],
      "metadata": {
        "id": "2se-A_I-ZN71"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dfSentimentStocks = dfSentimentStocks.ffill()"
      ],
      "metadata": {
        "id": "Y02Zi5-PY5EV"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dfSentimentStocks[\"comment_sentiment_average\"].plot()\n"
      ],
      "metadata": {
        "id": "PHdOYA89Y_N8",
        "outputId": "cc900cfb-9310-4778-9996-69cda7e3a5e0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 484
        }
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: xlabel='commentdate'>"
            ]
          },
          "metadata": {},
          "execution_count": 9
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}