NB_sentiment_model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/aibo/miniconda3/envs/python_env/lib/python3.10/site-packages/sklearn/metrics/_classification.py:2916: UserWarning: The y_pred values do not sum to one. Starting from 1.5 thiswill result in an error.\n",
      "  warnings.warn(\n",
      "/home/aibo/miniconda3/envs/python_env/lib/python3.10/site-packages/sklearn/metrics/_classification.py:2916: UserWarning: The y_pred values do not sum to one. Starting from 1.5 thiswill result in an error.\n",
      "  warnings.warn(\n",
      "/home/aibo/miniconda3/envs/python_env/lib/python3.10/site-packages/sklearn/metrics/_classification.py:2916: UserWarning: The y_pred values do not sum to one. Starting from 1.5 thiswill result in an error.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x800 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report:\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "    negative       0.83      0.82      0.82      2879\n",
      "    positive       0.88      0.89      0.89      4420\n",
      "\n",
      "    accuracy                           0.86      7299\n",
      "   macro avg       0.85      0.85      0.85      7299\n",
      "weighted avg       0.86      0.86      0.86      7299\n",
      "\n",
      "Accuracy: 0.8603918344978764\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix, ConfusionMatrixDisplay, log_loss\n",
    "import re\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "import joblib\n",
    "import gzip\n",
    "import pickle\n",
    "\n",
    "# Load and preprocess the data\n",
    "def preprocess_text(text):\n",
    "    # Case folding and normalization\n",
    "    text= str(text)\n",
    "    text = text.lower()\n",
    "    text = re.sub(r'[^a-zA-Z\\s]', '', text)\n",
    "    return text\n",
    "\n",
    "# Load dataset\n",
    "df = pd.read_csv('../DATASET/thesis_final_dataset.csv')\n",
    "\n",
    "# Apply text preprocessing\n",
    "df['cleaned_text'] = df['text'].apply(preprocess_text)\n",
    "\n",
    "# Split the dataset into training and testing sets\n",
    "train_df, test_df = train_test_split(df, test_size=0.3, random_state=42)\n",
    "\n",
    "# Encode labels\n",
    "le = LabelEncoder()\n",
    "le.fit(df['label'])  # Fit LabelEncoder to all labels in the dataset\n",
    "# df['label'] = le.fit_transform(df['label'])\n",
    "\n",
    "# Save the LabelEncoder to a pickle file\n",
    "with gzip.open('label_encoder.pkl', 'wb') as label_encoder_file:\n",
    "    pickle.dump(le, label_encoder_file)\n",
    "\n",
    "# Create Bag-of-Words representation using CountVectorizer\n",
    "vectorizer = CountVectorizer(max_features=5000)\n",
    "X_train = vectorizer.fit_transform(train_df['cleaned_text'])\n",
    "X_test = vectorizer.transform(test_df['cleaned_text'])\n",
    "y_train = le.transform(train_df['label'])  # Encode training labels\n",
    "y_test = le.transform(test_df['label'])  # Encode test labels\n",
    "\n",
    "# Check for unseen labels in test set and encode them\n",
    "unseen_labels = set(test_df['label'].unique()) - set(le.classes_)\n",
    "if unseen_labels:\n",
    "    print(\"Unseen labels in test set:\", unseen_labels)\n",
    "    # Encode unseen labels\n",
    "    test_df['label'] = test_df['label'].apply(lambda x: x if x in le.classes_ else 'other')\n",
    "\n",
    "# Remove instances with unseen labels from the test set\n",
    "test_df = test_df[test_df['label'].isin(le.classes_)]\n",
    "\n",
    "\n",
    "# Initialize Naive Bayes classifier\n",
    "nb_classifier = MultinomialNB()\n",
    "\n",
    "# Lists to store training and validation metrics\n",
    "train_loss_list, train_accuracy_list = [], []\n",
    "val_loss_list, val_accuracy_list = [], []\n",
    "misclassification_error_list = []\n",
    "\n",
    "# Split the training set into subsets\n",
    "subset_sizes = [int(len(train_df) * 0.2), int(len(train_df) * 0.5), len(train_df)]\n",
    "\n",
    "\n",
    "for subset_size in subset_sizes:\n",
    "    # Train Naive Bayes classifier on subset\n",
    "    nb_classifier.fit(X_train[:subset_size], y_train[:subset_size])\n",
    "\n",
    "    # Predictions on the validation set\n",
    "    y_val_pred = nb_classifier.predict(X_test)\n",
    "\n",
    "    # Calculate accuracy\n",
    "    val_accuracy = accuracy_score(y_test, y_val_pred)\n",
    "    val_loss = log_loss(y_test, nb_classifier.predict_proba(X_test))  # Calculate log-loss\n",
    "\n",
    "    # Append metrics to the lists\n",
    "    val_loss_list.append(val_loss)\n",
    "    val_accuracy_list.append(val_accuracy)\n",
    "\n",
    "    # Calculate misclassification error\n",
    "    misclassification_error = 1 - val_accuracy\n",
    "    misclassification_error_list.append(misclassification_error)\n",
    "\n",
    "# Plotting\n",
    "plt.figure(figsize=(20, 8))\n",
    "\n",
    "# Plot training and validation loss\n",
    "plt.subplot(1, 3, 1)\n",
    "plt.plot(subset_sizes, val_loss_list, label='Validation Log-loss', marker='o')\n",
    "plt.title('Validation Log-loss')\n",
    "plt.xlabel('Training Subset Size')\n",
    "plt.ylabel('Log-loss')\n",
    "plt.legend()\n",
    "\n",
    "# Plot training and validation accuracy\n",
    "plt.subplot(1, 3, 2)\n",
    "plt.plot(subset_sizes, val_accuracy_list, label='Validation Accuracy', marker='o')\n",
    "plt.title('Validation Accuracy')\n",
    "plt.xlabel('Training Subset Size')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "\n",
    "# Plot misclassification error\n",
    "plt.subplot(1, 3, 3)\n",
    "plt.plot(subset_sizes, misclassification_error_list, label='Misclassification Error', marker='o')\n",
    "plt.title('Misclassification Error')\n",
    "plt.xlabel('Training Subset Size')\n",
    "# plt.ylabel('Error Rate')\n",
    "plt.legend()\n",
    "\n",
    "# Generate and plot confusion matrix\n",
    "# plt.subplot(1, 4, 4)\n",
    "conf_matrix = confusion_matrix(y_test, y_val_pred)\n",
    "disp = ConfusionMatrixDisplay(conf_matrix, display_labels=le.classes_)\n",
    "disp.plot(cmap='Blues', values_format='d')\n",
    "plt.title('Confusion Matrix')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "# Generate classification report\n",
    "classification_report_result = classification_report(y_test, y_val_pred, target_names=le.classes_)\n",
    "print(\"Classification Report:\\n\", classification_report_result)\n",
    "\n",
    "# Evaluate the model on the test set\n",
    "y_pred = nb_classifier.predict(X_test)\n",
    "\n",
    "# Calculate accuracy\n",
    "accuracy = accuracy_score(y_test, y_pred)\n",
    "print('Accuracy:', accuracy)\n",
    "\n",
    "# # Save the trained model\n",
    "# joblib.dump(nb_classifier, \"naive_bayes_sentiment_model.pkl\")\n",
    "\n",
    "# Save the trained model\n",
    "with gzip.open('naive_bayes_sentiment_model.pkl', 'wb') as model_file:\n",
    "    pickle.dump(nb_classifier, model_file)\n",
    "\n",
    "# Save the CountVectorizer\n",
    "with gzip.open('vectorizer.pkl', 'wb') as vectorizer_file:\n",
    "    pickle.dump(vectorizer, vectorizer_file)\n",
    "\n",
    "# Save classification report to CSV\n",
    "classification_report_df = pd.DataFrame.from_dict(classification_report(y_test, y_pred, target_names=le.classes_, output_dict=True))\n",
    "classification_report_df.to_csv(\"../_classification_reports/NB_classification_report.csv\", mode='w', header=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of entries in the training dataframe: 3\n"
     ]
    }
   ],
   "source": [
    "train_count = len(subset_sizes)\n",
    "print(\"Number of entries in the training dataframe:\", train_count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[3406, 8515, 17031]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "subset_sizes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted labels for new texts: ['negative' 'negative' 'negative' 'positive' 'positive' 'negative'\n",
      " 'negative' 'positive' 'positive' 'positive' 'positive' 'negative'\n",
      " 'positive' 'positive' 'negative' 'positive' 'negative' 'positive'\n",
      " 'negative' 'negative' 'positive' 'negative']\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix, ConfusionMatrixDisplay, log_loss\n",
    "import re\n",
    "import numpy as np\n",
    "import gzip\n",
    "import pickle\n",
    "\n",
    "# Load and preprocess the data\n",
    "def preprocess_text(text):\n",
    "    # Case folding and normalization\n",
    "    text= str(text)\n",
    "    text = text.lower()\n",
    "    text = re.sub(r'[^a-zA-Z\\s]', '', text)\n",
    "    return text\n",
    "\n",
    "# Load LabelEncoder\n",
    "with gzip.open('label_encoder.pkl', 'rb') as label_encoder_file:\n",
    "    le = pickle.load(label_encoder_file)\n",
    "\n",
    "# Load the trained Naive Bayes classifier\n",
    "with gzip.open('naive_bayes_sentiment_model.pkl', 'rb') as model_file:\n",
    "    nb_classifier = pickle.load(model_file)\n",
    "\n",
    "# Load the CountVectorizer\n",
    "with gzip.open('vectorizer.pkl', 'rb') as vectorizer_file:\n",
    "    vectorizer = pickle.load(vectorizer_file)\n",
    "\n",
    "# Now you can use nb_classifier and vectorizer for classification tasks\n",
    "\n",
    "# List of comments for classification\n",
    "sample_comments = [\n",
    "    \"The disinterested teaching style of the instructor made it hard to fully comprehend and engage with the material\",\n",
    "    \"Hindi ko matukoy kung paano mapapakinabangan ang mga kasanayang ito sa totoong buhay\",\n",
    "    \"The course lacks real-world applications of machine learning that would enhance practical understanding.\",\n",
    "    \"your positivity is like a ray of sunshine on a cloudy day.\",\n",
    "    \"I'm grateful for the positive impact you've had on my education\",\n",
    "    \"The instructors' enthusiasm creates a positive learning environment where everyone feels encouraged to participate and ask questions\",\n",
    "    \"Hindi ako nakatutok sa lecture na ito\",\n",
    "    \"You show the true value of education.\",\n",
    "    \"Ipinapakita mo ang halaga ng pagiging positibo at pagiging bukas sa pagbabago sa aming mga buhay\",\n",
    "    \"You give meaning to our dreams.\",\n",
    "    \"Your class has ignited a passion for the subject in me\",\n",
    "    \"I didn't find the coursework challenging or stimulating\",\n",
    "    \"Napakahusay mong magbigay ng mga halimbawa na nagpapakita ng tunay na buhay na aplikasyon ng aming natutunan\",\n",
    "    \"You've provided valuable insights that will stay with me\",\n",
    "    \"I hoped for more enthusiasm and passion from our instructors\",\n",
    "    \"Your lessons shed light on our minds.\",\n",
    "    \"The instructor's lack of enthusiasm is reflected in the students' lack of interest\",\n",
    "    \"your perseverance in the face of challenges is truly admirable.\",\n",
    "    \"Minsan nakakalito ang pagkasunod-sunod ng mga topics\",\n",
    "    \"hindi mo maasahan sa bawat tanong\",\n",
    "    \"hindi sobrang magaling magturo si sir\",\n",
    "    \"not so bad, he teaches not very bad\",\n",
    "]\n",
    "\n",
    "# Apply text preprocessing to the sample comments\n",
    "sample_comments_preprocessed = [preprocess_text(comment) for comment in sample_comments]\n",
    "\n",
    "# Transform the sample comments using the same CountVectorizer\n",
    "sample_comments_transformed = vectorizer.transform(sample_comments_preprocessed)\n",
    "\n",
    "# Use the loaded model for inference\n",
    "predicted_labels = nb_classifier.predict(sample_comments_transformed)\n",
    "\n",
    "# Decode the predicted labels using the loaded label encoder\n",
    "decoded_labels = le.inverse_transform(predicted_labels)\n",
    "\n",
    "\n",
    "print(\"Predicted labels for new texts:\", decoded_labels)\n",
    "\n",
    "\n",
    "# predicted_sentiments = []\n",
    "# # Iterate over each text in the list\n",
    "# for text in sample_comments:\n",
    "#     # Preprocess the text\n",
    "#     preprocessed_text = preprocess_text(text)\n",
    "#     # Transform the preprocessed text using the loaded CountVectorizer\n",
    "#     vectorized_text = vectorizer.transform([preprocessed_text])\n",
    "#     # Predict the sentiment using the loaded Naive Bayes classifier\n",
    "#     predicted_sentiment = nb_classifier.predict(vectorized_text)\n",
    "#     # Map the predicted sentiment label back to its original category\n",
    "#     predicted_sentiment_label = le.inverse_transform([predicted_sentiment])\n",
    "#     # Append the predicted sentiment label to the list\n",
    "#     predicted_sentiments.append(predicted_sentiment_label)\n",
    "    \n",
    "#     # # Apply text preprocessing to the sample comments\n",
    "#     # sample_comments_preprocessed = [preprocess_text(comment) for comment in sample_comments]\n",
    "\n",
    "#     # # Transform the sample comments using the same CountVectorizer\n",
    "#     # sample_comments_transformed = vectorizer.transform(sample_comments_preprocessed)\n",
    "\n",
    "#     # # Use the loaded model for inference\n",
    "#     # predicted_labels = nb_classifier.predict(sample_comments_transformed)\n",
    "\n",
    "#     # # Decode the predicted labels using the loaded label encoder\n",
    "#     # decoded_labels = le.inverse_transform(predicted_labels)\n",
    "\n",
    "    \n",
    "    \n",
    "    \n",
    "\n",
    "# # # Print the predicted sentiments for each text\n",
    "# # for text, sentiment in zip(texts_to_classify, predicted_sentiments):\n",
    "# #     print(\"Text:\", text)\n",
    "# #     print(\"Predicted Sentiment:\", sentiment)\n",
    "# #     print()\n",
    "\n",
    "# neg_count = 0\n",
    "# pos_count = 0\n",
    "\n",
    "# # Display the results\n",
    "# for comment, sentiment in zip(sample_comments, predicted_sentiments):\n",
    "#     print(f\"Comment: {comment}\\nPredicted Sentiment: {sentiment}\\n{'='*50}\")\n",
    "#     if sentiment == \"negative\": neg_count+=1\n",
    "#     else: pos_count +=1\n",
    "\n",
    "# print(f\"pos_count {pos_count} \\nneg_count {neg_count}\")\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "python_env",
   "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": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}