anime AIU machine learning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b869aeb1",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\asus\\anaconda3\\lib\\site-packages\\scipy\\__init__.py:155: UserWarning: A NumPy version >=1.18.5 and <1.25.0 is required for this version of SciPy (detected version 1.26.4\n",
      "  warnings.warn(f\"A NumPy version >={np_minversion} and <{np_maxversion}\"\n",
      "C:\\Users\\asus\\AppData\\Local\\Temp\\ipykernel_15580\\1986867300.py:100: DeprecationWarning: ANTIALIAS is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.LANCZOS instead.\n",
      "  wallpaper = wallpaper.resize((1920, 1080), Image.ANTIALIAS)\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\n",
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "from sklearn.neighbors import NearestNeighbors\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "import tkinter as tk\n",
    "from tkinter import ttk\n",
    "from PIL import ImageTk, Image\n",
    "import requests\n",
    "from io import BytesIO\n",
    "\n",
    "# Data collection\n",
    "df = pd.read_csv('anime_list.csv')\n",
    "\n",
    "# Data preprocessing\n",
    "df = df.drop_duplicates(subset='title', keep='first')\n",
    "df = df.dropna(subset=['synopsis'])\n",
    "\n",
    "# Feature extraction with CountVectorizer\n",
    "cv = CountVectorizer(stop_words='english')\n",
    "synopsis_matrix = cv.fit_transform(df['synopsis'])\n",
    "cosine_sim_count = cosine_similarity(synopsis_matrix)\n",
    "\n",
    "# Feature extraction with TfidfVectorizer\n",
    "tfidf = TfidfVectorizer(stop_words='english')\n",
    "synopsis_matrix_tfidf = tfidf.fit_transform(df['synopsis'])\n",
    "cosine_sim_tfidf = cosine_similarity(synopsis_matrix_tfidf)\n",
    "\n",
    "# K-Nearest Neighbors model training with CountVectorizer\n",
    "knn = NearestNeighbors(metric='cosine', algorithm='brute')\n",
    "knn.fit(synopsis_matrix)\n",
    "\n",
    "# Random Forest model training with CountVectorizer\n",
    "rf = RandomForestRegressor(n_estimators=100, random_state=42)\n",
    "rf.fit(synopsis_matrix, df['score'])\n",
    "\n",
    "# Naive Bayes model training with CountVectorizer\n",
    "nb = MultinomialNB()\n",
    "nb.fit(synopsis_matrix, df['title'])\n",
    "\n",
    "# Function to get recommendation using cosine similarity with CountVectorizer\n",
    "def recommend_anime_count(title):\n",
    "    idx = df[df['title'] == title].index[0]\n",
    "    sim_scores = list(enumerate(cosine_sim_count[idx]))\n",
    "    sim_scores = sorted(sim_scores, key=lambda x: x[1], reverse=True)\n",
    "    sim_scores = sim_scores[1:11]\n",
    "    anime_indices = [i[0] for i in sim_scores]\n",
    "    return df['title'].iloc[anime_indices]\n",
    "\n",
    "# Function to get recommendation using cosine similarity with TfidfVectorizer\n",
    "def recommend_anime_tfidf(title):\n",
    "    idx = df[df['title'] == title].index[0]\n",
    "    sim_scores = list(enumerate(cosine_sim_tfidf[idx]))\n",
    "    sim_scores = sorted(sim_scores, key=lambda x: x[1], reverse=True)\n",
    "    sim_scores = sim_scores[1:11]\n",
    "    anime_indices = [i[0] for i in sim_scores]\n",
    "    return df['title'].iloc[anime_indices]\n",
    "\n",
    "# Function to get KNN recommendations\n",
    "def recommend_anime_knn(title):\n",
    "    idx = df[df['title'] == title].index[0]\n",
    "    distances, indices = knn.kneighbors(synopsis_matrix[idx], n_neighbors=11)\n",
    "    anime_indices = indices[0][1:]\n",
    "    return df['title'].iloc[anime_indices]\n",
    "\n",
    "# Function to get Random Forest recommendations\n",
    "def recommend_anime_rf(title):\n",
    "    idx = df[df['title'] == title].index[0]\n",
    "    synopsis = synopsis_matrix_tfidf[idx].toarray()\n",
    "    for i in range(10):\n",
    "        distances, indices = knn.kneighbors(synopsis, n_neighbors=11)\n",
    "        anime_indices = indices[0][1:]\n",
    "        synopsis = np.concatenate((synopsis, synopsis_matrix_tfidf[anime_indices].toarray()), axis=0)\n",
    "    scores = rf.predict(synopsis)\n",
    "    anime_indices = np.argsort(scores)[::-1][:10]\n",
    "    return df['title'].iloc[anime_indices]\n",
    "\n",
    "# Function to get Naive Bayes recommendations\n",
    "def recommend_anime_nb(title):\n",
    "    idx = df[df['title'] == title].index[0]\n",
    "    synopsis = synopsis_matrix_tfidf[idx].toarray()\n",
    "    for i in range(10):\n",
    "        distances, indices = knn.kneighbors(synopsis, n_neighbors=11)\n",
    "        anime_indices = indices[0][1:]\n",
    "        synopsis = np.concatenate((synopsis, synopsis_matrix_tfidf[anime_indices].toarray()), axis=0)\n",
    "    scores = nb.predict_proba(synopsis)\n",
    "    anime_indices = np.argsort(scores[:, 1])[::-1][:10]\n",
    "    return df['title'].iloc[anime_indices]\n",
    "\n",
    "# GUI setup\n",
    "root = tk.Tk()\n",
    "root.title('Anime Recommender')\n",
    "\n",
    "# Add wallpaper\n",
    "response = requests.get('https://th.bing.com/th/id/R.9507efa52b14e038eefc85ba5d74e448?rik=As6NiN0V3HtpZg&pid=ImgRaw&r=0')\n",
    "wallpaper = Image.open(BytesIO(response.content))\n",
    "wallpaper = wallpaper.resize((1920, 1080), Image.ANTIALIAS)\n",
    "wallpaper = ImageTk.PhotoImage(wallpaper)\n",
    "background_label = tk.Label(root, image=wallpaper)\n",
    "background_label.place(x=0, y=0, relwidth=1, relheight=1)\n",
    "\n",
    "# Label for title\n",
    "title_label = tk.Label(root, text='Anime Recommender', font=('Arial', 24))\n",
    "title_label.pack(pady=10)\n",
    "\n",
    "# Dropdown menu for selecting recommendation model\n",
    "model_label = tk.Label(root, text='Select a recommendation model:', font=('Arial', 14))\n",
    "model_label.pack()\n",
    "model_var = tk.StringVar()\n",
    "model_dropdown = ttk.Combobox(root, textvariable=model_var, font=('Arial', 14), state='readonly')\n",
    "model_dropdown['values'] = ('Cosine Similarity with CountVectorizer', 'Cosine Similarity with TfidfVectorizer', \n",
    "                            'K-Nearest Neighbors with CountVectorizer', 'Random Forest with CountVectorizer', \n",
    "                            'Naive Bayes with CountVectorizer')\n",
    "model_dropdown.pack(pady=10)\n",
    "\n",
    "# Label and entry for input title\n",
    "input_label = tk.Label(root, text='Enter an anime title:', font=('Arial', 14))\n",
    "input_label.pack()\n",
    "input_var = tk.StringVar()\n",
    "input_entry = tk.Entry(root, textvariable=input_var, font=('Arial', 14))\n",
    "input_entry.pack(pady=10)\n",
    "\n",
    "# Function to get recommendations based on selected model and input title\n",
    "def get_recommendations():\n",
    "    title = input_var.get()\n",
    "    model = model_var.get()\n",
    "    if model == 'Cosine Similarity with CountVectorizer':\n",
    "        recommendations = recommend_anime_count(title)\n",
    "    elif model == 'Cosine Similarity with TfidfVectorizer':\n",
    "        recommendations = recommend_anime_tfidf(title)\n",
    "    elif model == 'K-Nearest Neighbors with CountVectorizer':\n",
    "        recommendations = recommend_anime_knn(title)\n",
    "    elif model == 'Random Forest with CountVectorizer':\n",
    "        recommendations = recommend_anime_rf(title)\n",
    "    elif model == 'Naive Bayes with CountVectorizer':\n",
    "        recommendations = recommend_anime_nb(title)\n",
    "    output_text.delete('1.0', tk.END)\n",
    "    output_text.insert(tk.END, recommendations.to_string(index=False))\n",
    "\n",
    "# Button to get recommendations\n",
    "button = tk.Button(root, text='Get Recommendations', font=('Arial', 14), command=get_recommendations)\n",
    "button.pack(pady=10)\n",
    "\n",
    "# Output text area for recommendations\n",
    "output_frame = tk.Frame(root, bg='black')\n",
    "output_frame.pack(pady=10)\n",
    "output_label = tk.Label(output_frame, text='Recommendations:', font=('Arial', 14), fg='white', bg='black')\n",
    "output_label.pack(pady=5)\n",
    "output_text = tk.Text(output_frame, font=('Arial', 14))\n",
    "output_text.pack(pady=5)\n",
    "output_text.config(width=40, height=11)\n",
    "\n",
    "root.mainloop()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "39520e7a",
   "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}