{ "cells": [ { "cell_type": "markdown", "id": "57c229bb-bcf4-454d-adfc-8f93b8f329b1", "metadata": {}, "source": [ "Practical_Image_Processing_and_Natural_Language_Processing_First \n", "\n", "Author: Anwar Aba alwakl\n" ] }, { "cell_type": "markdown", "id": "63d8f882-0c0e-42d2-b345-3c95d3703eb7", "metadata": {}, "source": [ "1. Import Libraries" ] }, { "cell_type": "code", "execution_count": 29, "id": "b7609b39-a4bc-4781-9355-d449feb7944e", "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "import re\n", "\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.feature_extraction.text import TfidfVectorizer\n", "\n", "from sklearn.linear_model import LogisticRegression\n", "from sklearn.naive_bayes import MultinomialNB\n", "from sklearn.svm import LinearSVC\n", "\n", "from sklearn.metrics import accuracy_score, f1_score, classification_report" ] }, { "cell_type": "markdown", "id": "2596c871-b8f3-40dd-8ced-e44e4797c9d8", "metadata": {}, "source": [ "2. Load the Dataset" ] }, { "cell_type": "code", "execution_count": 30, "id": "4dd31f48-887d-40bb-bf69-fde2b5ae9421", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " text sentiment\n", "0 Java Concurrency in Practice is probably the b... positive\n", "1 haha aww hun i bet you are more creative tha... positive\n", "2 _pickle lol, thank you very much Hope you`re h... positive\n", "3 Out for an evening on the town with jeremy. Sa... negative\n", "4 - just took over the #1 Most Endorsed spot on... positive\n", "sentiment\n", "positive 2000\n", "negative 600\n", "Name: count, dtype: int64\n" ] } ], "source": [ "df = pd.read_csv(\"dataset.csv\")\n", "\n", "print(df.head())\n", "print(df['sentiment'].value_counts()) " ] }, { "cell_type": "markdown", "id": "1b541da8-61b7-4f4f-800d-817f7876150a", "metadata": {}, "source": [ "3. Data Preprocessing" ] }, { "cell_type": "code", "execution_count": 32, "id": "b451986e-f489-4192-b42a-f2ee7543b59f", "metadata": {}, "outputs": [], "source": [ "def clean_text(text):\n", " text = text.lower() \n", " text = re.sub(r'\\W', ' ', text) \n", " text = re.sub(r'\\s+', ' ', text) \n", " return text\n", "\n", "df['text'] = df['text'].apply(clean_text)\n", "\n", "\n", "df = df[['text', 'sentiment']].dropna()" ] }, { "cell_type": "markdown", "id": "12808f71-74e1-42a3-8bdc-d558b5eaf6ba", "metadata": {}, "source": [ "4) Split the Dataset" ] }, { "cell_type": "code", "execution_count": 43, "id": "3e923c7e-4b3f-4036-89f7-c5478182690f", "metadata": {}, "outputs": [], "source": [ "from sklearn.model_selection import train_test_split\n", "\n", "X_train, X_test, y_train, y_test = train_test_split(\n", " df['text'],\n", " df['sentiment'],\n", " test_size=0.2,\n", " stratify=df['sentiment'], \n", " random_state=42\n", ")" ] }, { "cell_type": "markdown", "id": "c5c2d278-19d8-4976-b656-04c99621ca52", "metadata": {}, "source": [ "5. Feature Representation (TF-IDF)" ] }, { "cell_type": "code", "execution_count": 44, "id": "c9766242-c042-4b2c-b254-d8238c1aa762", "metadata": {}, "outputs": [], "source": [ "from sklearn.feature_extraction.text import TfidfVectorizer\n", "\n", "vectorizer = TfidfVectorizer(max_features=5000)\n", "\n", "X_train_vec = vectorizer.fit_transform(X_train)\n", "X_test_vec = vectorizer.transform(X_test)" ] }, { "cell_type": "markdown", "id": "0d80ed5d-1d49-4d5f-8f52-47eda737a732", "metadata": {}, "source": [ "6) Train Models\n" ] }, { "cell_type": "markdown", "id": "34394e34-455a-4410-b805-c1ccdb708dd7", "metadata": {}, "source": [ " ✔️Logistic Regression ( imbalance handling)" ] }, { "cell_type": "code", "execution_count": 40, "id": "3da04afd-cec0-4ce1-b49d-7fd9d3954e56", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Accuracy: 0.8288461538461539\n", "F1: 0.8341589025703451\n", " precision recall f1-score support\n", "\n", " negative 0.61 0.73 0.66 120\n", " positive 0.91 0.86 0.89 400\n", "\n", " accuracy 0.83 520\n", " macro avg 0.76 0.80 0.77 520\n", "weighted avg 0.84 0.83 0.83 520\n", "\n" ] }, { "data": { "text/html": [ "
LogisticRegression(class_weight='balanced', max_iter=200)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "LogisticRegression(class_weight='balanced', max_iter=200)" ] }, "execution_count": 40, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y_pred = lr.predict(X_test_vec)\n", "\n", "print(\"Accuracy:\", accuracy_score(y_test, y_pred))\n", "print(\"F1:\", f1_score(y_test, y_pred, average='weighted'))\n", "print(classification_report(y_test, y_pred))\n", "lr = LogisticRegression(class_weight='balanced', max_iter=200)\n", "lr.fit(X_train_vec, y_train)" ] }, { "cell_type": "markdown", "id": "2c3a0641-701b-4201-91fe-de783fd72c92", "metadata": {}, "source": [ "✔️Naive Bayes" ] }, { "cell_type": "code", "execution_count": 41, "id": "7953edad-f235-4e8c-bf59-fbe72c0ccc5e", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Accuracy: 0.775\n", "F1 Score: 0.682341676197569\n", " precision recall f1-score support\n", "\n", " negative 1.00 0.03 0.05 120\n", " positive 0.77 1.00 0.87 400\n", "\n", " accuracy 0.78 520\n", " macro avg 0.89 0.51 0.46 520\n", "weighted avg 0.83 0.78 0.68 520\n", "\n" ] }, { "data": { "text/html": [ "
MultinomialNB()
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "MultinomialNB()" ] }, "execution_count": 41, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y_pred = nb.predict(X_test_vec)\n", "\n", "print(\"Accuracy:\", accuracy_score(y_test, y_pred))\n", "print(\"F1 Score:\", f1_score(y_test, y_pred, average='weighted'))\n", "print(classification_report(y_test, y_pred))\n", "nb = MultinomialNB()\n", "nb.fit(X_train_vec, y_train)" ] }, { "cell_type": "markdown", "id": "e0dd5e23-bdf7-4f41-a6b4-f496b40beee2", "metadata": {}, "source": [ "✔️ SVM" ] }, { "cell_type": "code", "execution_count": 42, "id": "6cf6c48a-f462-40dd-b71e-76f655848315", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Accuracy: 0.8346153846153846\n", "F1 Score: 0.8346153846153846\n", " precision recall f1-score support\n", "\n", " negative 0.64 0.64 0.64 120\n", " positive 0.89 0.89 0.89 400\n", "\n", " accuracy 0.83 520\n", " macro avg 0.77 0.77 0.77 520\n", "weighted avg 0.83 0.83 0.83 520\n", "\n" ] }, { "data": { "text/html": [ "
LinearSVC(class_weight='balanced')
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "LinearSVC(class_weight='balanced')" ] }, "execution_count": 42, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y_pred = svm.predict(X_test_vec)\n", "\n", "print(\"Accuracy:\", accuracy_score(y_test, y_pred))\n", "print(\"F1 Score:\", f1_score(y_test, y_pred, average='weighted'))\n", "print(classification_report(y_test, y_pred))\n", "svm = LinearSVC(class_weight='balanced')\n", "svm.fit(X_train_vec, y_train)" ] }, { "cell_type": "markdown", "id": "1be5dc27-829a-4565-8a1f-accb4e6d6d6d", "metadata": {}, "source": [ "7) Evaluation" ] }, { "cell_type": "code", "execution_count": 38, "id": "86853c15-dc78-4caa-9da7-0bf42233ee37", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "Logistic Regression\n", "Accuracy: 0.8288461538461539\n", "F1 Score: 0.8341589025703451\n", " precision recall f1-score support\n", "\n", " negative 0.61 0.73 0.66 120\n", " positive 0.91 0.86 0.89 400\n", "\n", " accuracy 0.83 520\n", " macro avg 0.76 0.80 0.77 520\n", "weighted avg 0.84 0.83 0.83 520\n", "\n", "\n", "Naive Bayes\n", "Accuracy: 0.775\n", "F1 Score: 0.682341676197569\n", " precision recall f1-score support\n", "\n", " negative 1.00 0.03 0.05 120\n", " positive 0.77 1.00 0.87 400\n", "\n", " accuracy 0.78 520\n", " macro avg 0.89 0.51 0.46 520\n", "weighted avg 0.83 0.78 0.68 520\n", "\n", "\n", "SVM\n", "Accuracy: 0.8346153846153846\n", "F1 Score: 0.8346153846153846\n", " precision recall f1-score support\n", "\n", " negative 0.64 0.64 0.64 120\n", " positive 0.89 0.89 0.89 400\n", "\n", " accuracy 0.83 520\n", " macro avg 0.77 0.77 0.77 520\n", "weighted avg 0.83 0.83 0.83 520\n", "\n" ] } ], "source": [ "def evaluate(model, X_test, y_test, name):\n", " y_pred = model.predict(X_test)\n", " print(f\"\\n{name}\")\n", " print(\"Accuracy:\", accuracy_score(y_test, y_pred))\n", " print(\"F1 Score:\", f1_score(y_test, y_pred, average='weighted'))\n", " print(classification_report(y_test, y_pred))\n", "\n", "evaluate(lr, X_test_vec, y_test, \"Logistic Regression\")\n", "evaluate(nb, X_test_vec, y_test, \"Naive Bayes\")\n", "evaluate(svm, X_test_vec, y_test, \"SVM\")" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.13.5" } }, "nbformat": 4, "nbformat_minor": 5 }