{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [
        "zUoqlTw3ECKM",
        "Nrfoc5dTGLWg",
        "mr0oOrO7Imoi"
      ]
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Word Embeddings with gensim\n",
        "\n",
        "Notebook Contents:\n",
        "- Load word embeddings using [gensim](https://radimrehurek.com/gensim/) package\n",
        "- Calculate word similarity\n",
        "- Train a word2vec model using gensim"
      ],
      "metadata": {
        "id": "C4qEG47MDHeQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import logging\n",
        "logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)"
      ],
      "metadata": {
        "id": "tywE5kJPFqhg"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Load pre-trained word2vec model"
      ],
      "metadata": {
        "id": "zUoqlTw3ECKM"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import gensim.downloader as api\n",
        "wv = api.load('glove-wiki-gigaword-300')"
      ],
      "metadata": {
        "id": "CHgemDH5FmYk",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "83da8f81-0f59-49ff-de55-6eeda15eea0a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[==================================================] 100.0% 376.1/376.1MB downloaded\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "len(wv)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Fov4PP5e7nQd",
        "outputId": "005ea926-e060-401b-f677-001568d44002"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "400000"
            ]
          },
          "metadata": {},
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We can obtain vector representation of a word"
      ],
      "metadata": {
        "id": "hT5DXTA5Fvf3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "vec_king = wv['king']\n",
        "vec_king.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OWUyPVsuGE3D",
        "outputId": "0a2ee146-b66c-416d-fc8a-d77dcfea488d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(300,)"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Calculate word similarity\n",
        "\n",
        "Using the function `wv.similarity`"
      ],
      "metadata": {
        "id": "Nrfoc5dTGLWg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "pairs = [\n",
        "    ('car', 'minivan'),   # a minivan is a kind of car\n",
        "    ('car', 'automobile'),   # a minivan is a kind of car\n",
        "    ('car', 'bicycle'),   # still a wheeled vehicle\n",
        "    ('car', 'airplane'),  # ok, no wheels, but still a vehicle\n",
        "    ('car', 'cereal'),    # ... and so on\n",
        "    ('car', 'communism'),\n",
        "]\n",
        "for w1, w2 in pairs:\n",
        "    print('%r\\t%r\\t%.2f' % (w1, w2, wv.similarity(w1, w2)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sTPXuFW7GPlm",
        "outputId": "1dd9eea2-3738-41ab-e076-be4efe04ab35"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "'car'\t'minivan'\t0.50\n",
            "'car'\t'automobile'\t0.60\n",
            "'car'\t'bicycle'\t0.50\n",
            "'car'\t'airplane'\t0.43\n",
            "'car'\t'cereal'\t0.03\n",
            "'car'\t'communism'\t0.02\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Print the 5 most similar words to “car” or “minivan”"
      ],
      "metadata": {
        "id": "qiRSPcqmIBY2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print(wv.most_similar(positive=['car', 'minivan'], topn=5))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QXqfeOvqIDvS",
        "outputId": "1ac2e702-6503-4363-954b-b96b9beaf120"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[('suv', 0.7696972489356995), ('vehicle', 0.7469112873077393), ('truck', 0.7312718629837036), ('cars', 0.7033854722976685), ('jeep', 0.6848679184913635)]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Which of the below does not belong in the sequence?"
      ],
      "metadata": {
        "id": "YZeTiAKxIHc9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print(wv.doesnt_match(['fire', 'water', 'land', 'sea', 'air', 'car']))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0HK0as1TIN67",
        "outputId": "bbc0ff79-a1cd-4d70-f783-a060fb0e67fe"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "car\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Word Analogy\n"
      ],
      "metadata": {
        "id": "mr0oOrO7Imoi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print(wv.similar_by_vector(wv['spain'] - wv['madrid'] + wv['athens'], topn=10))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "___wFYKdJAeO",
        "outputId": "d8705063-6c25-47f8-fd1f-c56cdb365a65"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[('greece', 0.7637240886688232), ('athens', 0.7158880233764648), ('spain', 0.5469861626625061), ('greek', 0.5434280633926392), ('cyprus', 0.507988452911377), ('bulgaria', 0.49355757236480713), ('portugal', 0.4708733856678009), ('hungary', 0.4684614837169647), ('crete', 0.4490693211555481), ('greeks', 0.4459525942802429)]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(wv.similar_by_vector(wv['king'] - wv['man'] + wv['woman'], topn=10))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LY_zN2X2kApQ",
        "outputId": "4c9e4844-5357-4440-f05f-f9da4a5b6fa2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[('king', 0.8065858483314514), ('queen', 0.689616322517395), ('monarch', 0.5575491189956665), ('throne', 0.5565375089645386), ('princess', 0.5518684387207031), ('mother', 0.5142154693603516), ('daughter', 0.5133156776428223), ('kingdom', 0.5025345087051392), ('prince', 0.5017741322517395), ('elizabeth', 0.4908031225204468)]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Train word2vec model\n",
        "Dataset: [Coronavirus Tweets Dataset](https://www.kaggle.com/datatattle/covid-19-nlp-text-classification)"
      ],
      "metadata": {
        "id": "_FQuQdilrxQ9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "! pip install kaggle"
      ],
      "metadata": {
        "id": "GqcsQJz6wxde",
        "outputId": "186195c3-4937-4e61-920a-e4f2d8079089",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: kaggle in /usr/local/lib/python3.10/dist-packages (1.6.17)\n",
            "Requirement already satisfied: six>=1.10 in /usr/local/lib/python3.10/dist-packages (from kaggle) (1.17.0)\n",
            "Requirement already satisfied: certifi>=2023.7.22 in /usr/local/lib/python3.10/dist-packages (from kaggle) (2024.12.14)\n",
            "Requirement already satisfied: python-dateutil in /usr/local/lib/python3.10/dist-packages (from kaggle) (2.8.2)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from kaggle) (2.32.3)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from kaggle) (4.67.1)\n",
            "Requirement already satisfied: python-slugify in /usr/local/lib/python3.10/dist-packages (from kaggle) (8.0.4)\n",
            "Requirement already satisfied: urllib3 in /usr/local/lib/python3.10/dist-packages (from kaggle) (2.2.3)\n",
            "Requirement already satisfied: bleach in /usr/local/lib/python3.10/dist-packages (from kaggle) (6.2.0)\n",
            "Requirement already satisfied: webencodings in /usr/local/lib/python3.10/dist-packages (from bleach->kaggle) (0.5.1)\n",
            "Requirement already satisfied: text-unidecode>=1.3 in /usr/local/lib/python3.10/dist-packages (from python-slugify->kaggle) (1.3)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->kaggle) (3.4.0)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->kaggle) (3.10)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!rm -f covid-19-nlp-text-classification.zip\n",
        "!rm -f Corona_NLP_test.csv\n",
        "!rm -f Corona_NLP_train.csv\n",
        "\n",
        "!kaggle datasets download datatattle/covid-19-nlp-text-classification\n",
        "!unzip covid-19-nlp-text-classification.zip"
      ],
      "metadata": {
        "id": "Ds1G5tAhwaoZ",
        "outputId": "0bffddfd-ef6f-4246-954f-fe78579a959b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Dataset URL: https://www.kaggle.com/datasets/datatattle/covid-19-nlp-text-classification\n",
            "License(s): copyright-authors\n",
            "Downloading covid-19-nlp-text-classification.zip to /content\n",
            "  0% 0.00/4.38M [00:00<?, ?B/s]\n",
            "100% 4.38M/4.38M [00:00<00:00, 190MB/s]\n",
            "Archive:  covid-19-nlp-text-classification.zip\n",
            "  inflating: Corona_NLP_test.csv     \n",
            "  inflating: Corona_NLP_train.csv    \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd"
      ],
      "metadata": {
        "id": "J0JLTTPFvmg1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dataset = pd.read_csv('Corona_NLP_train.csv', encoding='latin1')"
      ],
      "metadata": {
        "id": "fXA6LPdBvpHq"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Text preprocessing by removing:\n",
        "- punctuations, numbers\n",
        "- characters including and after 'https' in the text"
      ],
      "metadata": {
        "id": "hKk1kxJazKK5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import re\n",
        "texts=[]\n",
        "for i in range(0,len(dataset)):\n",
        "  text = re.sub('[^a-zA-Z]' , ' ', dataset['OriginalTweet'][i])\n",
        "  text = text.lower()\n",
        "  text = text.split()\n",
        "  x = len(text) if text.count('https') ==0  else text.index('https')\n",
        "  text = text[: x ]\n",
        "  text = [t for t in text if not t=='https']\n",
        "  text = ' '.join(text)\n",
        "  texts.append(text)\n",
        "print(len(texts))"
      ],
      "metadata": {
        "id": "sRJDe_sYzZgt",
        "outputId": "2563dc1a-bf90-4f45-c45b-1e14fb9d4a57",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "41157\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Training the word2vec model"
      ],
      "metadata": {
        "id": "j0ZIGwRCzgZ2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from gensim.models import Word2Vec\n",
        "\n",
        "sentences = [line.split() for line in texts]\n",
        "w2v = Word2Vec(sentences, vector_size=100, window=5, workers=4, min_count=5)"
      ],
      "metadata": {
        "id": "XrRkwKpNzlDR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "vocab = list(w2v.wv.key_to_index.keys())\n",
        "print(len(vocab))"
      ],
      "metadata": {
        "id": "ooFLva9OzjvR",
        "outputId": "37188bca-0fd1-47fa-9630-7c733d6a4c59",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "10630\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.decomposition import PCA\n",
        "import random\n",
        "\n",
        "def display_pca_scatterplot(model, word, sample=10, topn=5):\n",
        "    \"\"\"\n",
        "    Plot a word, its most similar words, and random sample words in 2D using PCA.\n",
        "    \"\"\"\n",
        "    sample_words = random.sample(vocab, sample) # random words\n",
        "    close_words = [w for w, _ in model.wv.most_similar([word], topn=topn)] # most similar words\n",
        "\n",
        "    words = np.concatenate(([word], close_words, sample_words))\n",
        "    word_vectors = np.array([model.wv[w] for w in words])\n",
        "\n",
        "    # Apply PCA\n",
        "    pca = PCA(n_components=2)\n",
        "    coords = pca.fit_transform(word_vectors)\n",
        "\n",
        "    # Plot setup\n",
        "    plt.figure(figsize=(10, 8))\n",
        "    colors = ['blue'] + ['green'] * topn + ['red'] * sample\n",
        "\n",
        "    # Plot points and annotations in one go\n",
        "    plt.scatter(coords[:, 0], coords[:, 1], c=colors)\n",
        "\n",
        "    for coord, word in zip(coords, words):\n",
        "        plt.annotate(word, xy=coord, fontsize=12, xytext=(5, 5),textcoords='offset points')\n",
        "\n",
        "    plt.title(\"Word Embeddings Visualization\")\n",
        "    plt.xlabel(\"PCA Component 1\")\n",
        "    plt.ylabel(\"PCA Component 2\")\n",
        "    plt.grid(True)\n",
        "    plt.tight_layout()\n",
        "    plt.show()"
      ],
      "metadata": {
        "id": "GNKBD0au1NWI"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "display_pca_scatterplot(w2v,'corona')"
      ],
      "metadata": {
        "id": "9UHhPe6wI4dw",
        "outputId": "5d5145ec-4e23-4490-8eb9-70e577ffd017",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 807
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x800 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Save our existing model and load it again."
      ],
      "metadata": {
        "id": "QS10bmCM28Gk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "w2v.save(\"word2vec.model\")\n",
        "model = Word2Vec.load(\"word2vec.model\")"
      ],
      "metadata": {
        "id": "7hH4VLZ827T1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## References\n",
        "\n",
        "- [Word2Vec Model Tutorial](https://radimrehurek.com/gensim/auto_examples/tutorials/run_word2vec.html#sphx-glr-auto-examples-tutorials-run-word2vec-py) on gensim documentation\n",
        "- [medium article](https://medium.com/@manansuri/a-dummys-guide-to-word2vec-456444f3c673)"
      ],
      "metadata": {
        "id": "hUQIMRBvFYA8"
      }
    }
  ]
}