{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "cba3bba5", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "[nltk_data] Downloading package punkt to\n", "[nltk_data] C:\\Users\\mmoghyiri\\AppData\\Roaming\\nltk_data...\n", "[nltk_data] Package punkt is already up-to-date!\n", "[nltk_data] Downloading package punkt_tab to\n", "[nltk_data] C:\\Users\\mmoghyiri\\AppData\\Roaming\\nltk_data...\n", "[nltk_data] Package punkt_tab is already up-to-date!\n", "[nltk_data] Downloading package stopwords to\n", "[nltk_data] C:\\Users\\mmoghyiri\\AppData\\Roaming\\nltk_data...\n", "[nltk_data] Package stopwords is already up-to-date!\n", "[nltk_data] Downloading package wordnet to\n", "[nltk_data] C:\\Users\\mmoghyiri\\AppData\\Roaming\\nltk_data...\n", "[nltk_data] Package wordnet is already up-to-date!\n", "[nltk_data] Downloading package punkt to\n", "[nltk_data] C:\\Users\\mmoghyiri\\AppData\\Roaming\\nltk_data...\n", "[nltk_data] Package punkt is already up-to-date!\n", "[nltk_data] Downloading package stopwords to\n", "[nltk_data] C:\\Users\\mmoghyiri\\AppData\\Roaming\\nltk_data...\n", "[nltk_data] Package stopwords is already up-to-date!\n", "[nltk_data] Downloading package wordnet to\n", "[nltk_data] C:\\Users\\mmoghyiri\\AppData\\Roaming\\nltk_data...\n", "[nltk_data] Package wordnet is already up-to-date!\n" ] }, { "data": { "text/plain": [ "True" ] }, "execution_count": 1, "metadata": {}, "output_type": "execute_result" } ], "source": [ "\n", "import pandas as pd\n", "import numpy as np\n", "import re\n", "import nltk\n", "import seaborn as sns\n", "import matplotlib.pyplot as plt\n", "\n", "from sklearn.model_selection import train_test_split, GridSearchCV\n", "from sklearn.feature_extraction.text import TfidfVectorizer, CountVectorizer\n", "from sklearn.linear_model import LogisticRegression\n", "from sklearn.naive_bayes import MultinomialNB\n", "from sklearn.ensemble import RandomForestClassifier\n", "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, classification_report, confusion_matrix\n", "\n", "from gensim.models import Word2Vec\n", "from nltk.tokenize import word_tokenize\n", "from nltk.corpus import stopwords\n", "from nltk.stem import PorterStemmer, WordNetLemmatizer\n", "import nltk\n", "for pkg in [\"punkt\", \"punkt_tab\", \"stopwords\", \"wordnet\"]:\n", " nltk.download(pkg)\n", "\n", "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, classification_report, confusion_matrix\n", "import seaborn as sns\n", "import matplotlib.pyplot as plt\n", " \n", "nltk.download('punkt')\n", "nltk.download('stopwords')\n", "nltk.download('wordnet')\n" ] }, { "cell_type": "code", "execution_count": 2, "id": "e65118fd", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " label message label_num\n", "0 ham Go until jurong point, crazy.. Available only ... 0\n", "1 ham Ok lar... Joking wif u oni... 0\n", "2 spam Free entry in 2 a wkly comp to win FA Cup fina... 1\n", "3 ham U dun say so early hor... U c already then say... 0\n", "4 ham Nah I don't think he goes to usf, he lives aro... 0\n", "5 spam FreeMsg Hey there darling it's been 3 week's n... 1\n", "6 ham Even my brother is not like to speak with me. ... 0\n", "7 ham As per your request 'Melle Melle (Oru Minnamin... 0\n", "8 spam WINNER!! As a valued network customer you have... 1\n", "9 spam Had your mobile 11 months or more? U R entitle... 1\n", "10 ham I'm gonna be home soon and i don't want to tal... 0\n", "11 spam SIX chances to win CASH! From 100 to 20,000 po... 1\n", "12 spam URGENT! You have won a 1 week FREE membership ... 1\n", "13 ham I've been searching for the right words to tha... 0\n", "14 ham I HAVE A DATE ON SUNDAY WITH WILL!! 0\n", "15 spam XXXMobileMovieClub: To use your credit, click ... 1\n", "16 ham Oh k...i'm watching here:) 0\n", "17 ham Eh u remember how 2 spell his name... Yes i di... 0\n", "18 ham Fine if thatåÕs the way u feel. ThatåÕs th... 0\n", "19 spam England v Macedonia - dont miss the goals/team... 1\n", "\n", "label\n", "ham 4825\n", "spam 747\n", "Name: count, dtype: int64\n", "\n", "label_num\n", "0 4825\n", "1 747\n", "Name: count, dtype: int64\n", "\n" ] } ], "source": [ "\n", "# load data\n", "\n", "df = pd.read_csv(\"spam_sms.csv\", encoding=\"latin-1\")\n", "df.columns = [\"label\", \"message\"]\n", "df['label_num'] = df['label'].map({'ham': 0, 'spam': 1})\n", "\n", "print(df.head(20), end=\"\\n\\n\")\n", "print(df['label'].value_counts(), end=\"\\n\\n\")\n", "print(df['label_num'].value_counts(), end=\"\\n\\n\")\n" ] }, { "cell_type": "code", "execution_count": null, "id": "1068e1c8-cc12-4388-a8f0-858f70d2bec6", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "f0aafdff-0420-44d3-9536-5fa4aa368d94", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 5, "id": "09913d2f", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " label_num label_num clean_text\n", "0 0 0 go jurong point crazy available bugis n great ...\n", "1 0 0 ok lar joking wif u oni\n", "2 1 1 free entry wkly comp win fa cup final tkts st ...\n", "3 0 0 u dun say early hor u c already say\n", "4 0 0 nah dont think go usf life around though\n", "5 1 1 freemsg hey darling week word back id like fun...\n", "6 0 0 even brother like speak treat like aid patent\n", "7 0 0 per request melle melle oru minnaminunginte nu...\n", "8 1 1 winner valued network customer selected receiv...\n", "9 1 1 mobile month u r entitled update latest colour...\n", "10 0 0 im gon na home soon dont want talk stuff anymo...\n", "11 1 1 six chance win cash pound txt csh send cost pd...\n", "12 1 1 urgent week free membership prize jackpot txt ...\n", "13 0 0 ive searching right word thank breather promis...\n", "14 0 0 date sunday\n", "15 1 1 xxxmobilemovieclub use credit click wap link n...\n", "16 0 0 oh kim watching\n", "17 0 0 eh u remember spell name yes v naughty make v wet\n", "18 0 0 fine thats way u feel thats way gota b\n", "19 1 1 england v macedonia dont miss goalsteam news t...\n", "20 0 0 seriously spell name\n", "21 0 0 im going try month ha ha joking\n", "22 0 0 pay first lar da stock comin\n", "23 0 0 aft finish lunch go str lor ard smth lor u fin...\n", "24 0 0 ffffffffff alright way meet sooner\n", "25 0 0 forced eat slice im really hungry tho suck mar...\n", "26 0 0 lol always convincing\n", "27 0 0 catch bus frying egg make tea eating mom left ...\n", "28 0 0 im back amp packing car ill let know there room\n", "29 0 0 ahhh work vaguely remember feel like lol\n", "30 0 0 wait thats still clear sure sarcastic thats x ...\n", "31 0 0 yeah got v apologetic n fallen actin like spoi...\n", "32 0 0 k tell anything\n", "33 0 0 fear fainting housework quick cuppa\n", "34 1 1 thanks subscription ringtone uk mobile charged...\n", "35 0 0 yup ok go home look timing msg xuhui going lea...\n", "36 0 0 oops ill let know roommate done\n", "37 0 0 see letter b car\n", "38 0 0 anything lor u decide\n", "39 0 0 hello hows saturday go texting see youd decide...\n", "40 0 0 pls go ahead watt wanted sure great weekend ab...\n", "41 0 0 forget tell want need crave love sweet arabian...\n", "42 1 1 rodger burn msg tried call reply sm free nokia...\n", "43 0 0 seeing\n", "44 0 0 great hope like man well endowed ltgt inch\n", "45 0 0 callsmessagesmissed call\n", "46 0 0 didnt get hep b immunisation nigeria\n", "47 0 0 fair enough anything going\n", "48 0 0 yeah hopefully tyler cant could maybe ask arou...\n", "49 0 0 u dont know stubborn didnt even want go hospit...\n" ] } ], "source": [ "# preprocessing\n", "\n", "stop_words = set(stopwords.words('english'))\n", "stemmer = PorterStemmer()\n", "lemmatizer = WordNetLemmatizer()\n", "\n", "def preprocess_text(text):\n", " text = text.lower()\n", " text = re.sub(r'http\\S+|www\\.\\S+', '', text)\n", " text = re.sub(r'<.*?>', '', text)\n", " text = re.sub(r'[^a-zA-Z\\s]', '', text)\n", " text = str(text).lower().strip()\n", " text = url_pat.sub('', text)\n", " text = html_pat.sub('', text)\n", " text = non_alpha_pat.sub(' ', text)\n", "\n", " tokens = word_tokenize(text, preserve_line=True)\n", " tokens = [t for t in tokens if t not in stop_words]\n", " tokens = [lemmatizer.lemmatize(t) for t in tokens]\n", " return ' '.join(tokens)\n", "\n", "df['clean_text'] = df['message'].apply(preprocess_text)\n", "print(df[['label_num','label_num', 'clean_text']].head(50))\n" ] }, { "cell_type": "code", "execution_count": null, "id": "aa6b8777", "metadata": {}, "outputs": [], "source": [ "# train & split\n", "\n", "X = df['clean_text']\n", "y = df['label_num']\n", "\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42, stratify=y)\n", "\n", "len(X_train), len(X_test), np.unique(y, return_counts=True)\n", "\n" ] }, { "cell_type": "code", "execution_count": null, "id": "ecb0ed18", "metadata": {}, "outputs": [], "source": [ "vectorizer = TfidfVectorizer(ngram_range=(1,2))\n", "X_train_vec = vectorizer.fit_transform(X_train)\n", "X_test_vec = vectorizer.transform(X_test)\n", "\n", "print(\"TF-IDF shape:\", X_train_vec.shape)\n" ] }, { "cell_type": "code", "execution_count": null, "id": "87c61a2e", "metadata": {}, "outputs": [], "source": [ "# evaluate\n", "\n", "def evaluate_model(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(\"Precision:\", precision_score(y_test, y_pred, zero_division=0))\n", " print(\"Recall:\", recall_score(y_test, y_pred, zero_division=0))\n", " print(\"F1:\", f1_score(y_test, y_pred, zero_division=0))\n", " print(\"\\nClassification Report:\\n\", classification_report(y_test, y_pred, zero_division=0))\n", " cm = confusion_matrix(y_test, y_pred)\n", " sns.heatmap(cm, annot=True, fmt=\"d\", cbar=False)\n", " plt.title(f\"Confusion Matrix - {name}\")\n", " plt.xlabel(\"Predicted\")\n", " plt.ylabel(\"Actual\")\n", " plt.show()\n", "\n", "lr = LogisticRegression(max_iter=200)\n", "lr.fit(X_train_vec, y_train)\n", "evaluate_model(lr, X_test_vec, y_test, \"Logistic Regression\")\n", "\n", "nb = MultinomialNB()\n", "nb.fit(X_train_vec, y_train)\n", "evaluate_model(nb, X_test_vec, y_test, \"Naive Bayes\")\n", "\n", "rf = RandomForestClassifier(n_estimators=200, max_depth=20, random_state=42)\n", "rf.fit(X_train_vec, y_train)\n", "evaluate_model(rf, X_test_vec, y_test, \"Random Forest\")\n" ] }, { "cell_type": "code", "execution_count": null, "id": "98cf45d7", "metadata": {}, "outputs": [], "source": [ "df['tokens'] = df['clean_text'].apply(lambda x: word_tokenize(x))\n", "\n", "w2v_model = Word2Vec(sentences=df['tokens'], vector_size=100, window=5, min_count=1, sg=0)\n", "\n", "def get_sentence_vector(tokens, model):\n", " vectors = [model.wv[w] for w in tokens if w in model.wv]\n", " return np.mean(vectors, axis=0) if vectors else np.zeros(model.vector_size)\n", "\n", "df['vector'] = df['tokens'].apply(lambda x: get_sentence_vector(x, w2v_model))\n", "\n", "X = np.vstack(df['vector'].values)\n", "y = df['label_num'].values\n", "\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42, stratify=y)\n", "\n", "lr_w2v = LogisticRegression(max_iter=200)\n", "lr_w2v.fit(X_train, y_train)\n", "evaluate_model(lr_w2v, X_test, y_test, \"Logistic Regression (Word2Vec)\")\n" ] }, { "cell_type": "code", "execution_count": null, "id": "f168ba8a-681f-4a47-a9f9-da37dcd6fe63", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "6878fff8-d5fe-4863-9a96-31a6740302f4", "metadata": {}, "outputs": [], "source": [] } ], "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": "3.11.13" } }, "nbformat": 4, "nbformat_minor": 5 }