diff --git a/models/similarity_matrix.pkl b/models/similarity_matrix.pkl
index d310511..addcefc 100644
Binary files a/models/similarity_matrix.pkl and b/models/similarity_matrix.pkl differ
diff --git a/notebooks/model_training.ipynb b/notebooks/model_training.ipynb
index 7860d40..56a7012 100644
--- a/notebooks/model_training.ipynb
+++ b/notebooks/model_training.ipynb
@@ -40,15 +40,12 @@
     }
    ],
    "source": [
-    "# Import Dependencies\n",
-    "\n",
     "import numpy as np\n",
     "import pandas as pd\n",
-    "import matplotlib.pyplot as plt\n",
-    "import seaborn as sns\n",
     "from sklearn.feature_extraction.text import TfidfVectorizer\n",
     "from sklearn.metrics.pairwise import cosine_similarity\n",
     "from nltk.stem import WordNetLemmatizer\n",
+    "from sklearn.decomposition import TruncatedSVD\n",
     "import pickle\n",
     "import nltk\n",
     "import re\n",
@@ -59,57 +56,10 @@
     "except LookupError:\n",
     "    nltk.download('wordnet')\n",
     "\n",
+    "\n",
     "print('Dependencies Imported')"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "id": "9cfffeab",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "/Users/bushra/data-science/upwork/education-projects/03/course-recommendation-system/notebooks\n"
-     ]
-    }
-   ],
-   "source": [
-    "import os\n",
-    "current_directory = os.getcwd()\n",
-    "print(current_directory)\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "id": "8fafd9bc",
-   "metadata": {
-    "scrolled": false
-   },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "/Users/bushra/data-science/upwork/education-projects/03/course-recommendation-system\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/Users/bushra/data-science/upwork/education-projects/03/course-recommendation-system/venv/lib/python3.11/site-packages/IPython/core/magics/osm.py:417: UserWarning: This is now an optional IPython functionality, setting dhist requires you to install the `pickleshare` library.\n",
-      "  self.shell.db['dhist'] = compress_dhist(dhist)[-100:]\n"
-     ]
-    }
-   ],
-   "source": [
-    "%cd .."
-   ]
-  },
   {
    "cell_type": "markdown",
    "id": "2b7a46c9",
@@ -120,29 +70,17 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 2,
    "id": "046017bb",
    "metadata": {},
    "outputs": [],
    "source": [
-    "import pandas as pd\n",
-    "\n",
-    "# Function to load CSV data\n",
-    "def load_data(file_path):\n",
-    "    try:\n",
-    "        data = pd.read_csv(file_path, encoding='utf-8')\n",
-    "        return data\n",
-    "    except FileNotFoundError:\n",
-    "        print(f\"File {file_path} not found.\")\n",
-    "        return None\n",
-    "\n",
-    "# Loading the data from a local file\n",
-    "data = load_data(\"../course-recommendation-system/data/coursera.csv\")\n"
+    "data = pd.read_csv(\"coursera.csv\", encoding='utf-8')"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 3,
    "id": "5fd1ffb0",
    "metadata": {},
    "outputs": [
@@ -268,7 +206,7 @@
        "4  Data Analysis  select (sql)  database manageme...  "
       ]
      },
-     "execution_count": 5,
+     "execution_count": 3,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -287,7 +225,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 4,
    "id": "5cb80836",
    "metadata": {
     "scrolled": true
@@ -299,19 +237,18 @@
        "(3522, 7)"
       ]
      },
-     "execution_count": 7,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "# Display dataset shape (number of rows and columns)\n",
     "data.shape"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 5,
    "id": "95cde0c6",
    "metadata": {
     "scrolled": true
@@ -339,205 +276,110 @@
     }
    ],
    "source": [
-    "# Show summary of the dataset (information on column types and memory usage)\n",
     "data.info()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 6,
    "id": "f08d2540",
    "metadata": {},
    "outputs": [
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Course Name           0\n",
-      "University            0\n",
-      "Difficulty Level      0\n",
-      "Course Rating         0\n",
-      "Course URL            0\n",
-      "Course Description    0\n",
-      "Skills                0\n",
-      "dtype: int64\n"
-     ]
+     "data": {
+      "text/plain": [
+       "Course Name           0\n",
+       "University            0\n",
+       "Difficulty Level      0\n",
+       "Course Rating         0\n",
+       "Course URL            0\n",
+       "Course Description    0\n",
+       "Skills                0\n",
+       "dtype: int64"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [
-    "# Check for missing values in each column\n",
-    "print(data.isnull().sum())"
+    "data.isnull().sum()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 7,
    "id": "01cab0bc",
    "metadata": {
     "scrolled": true
    },
    "outputs": [
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Course Name           3416\n",
-      "University             184\n",
-      "Difficulty Level         5\n",
-      "Course Rating           31\n",
-      "Course URL            3424\n",
-      "Course Description    3397\n",
-      "Skills                3424\n",
-      "dtype: int64\n"
-     ]
+     "data": {
+      "text/plain": [
+       "Course Name           3416\n",
+       "University             184\n",
+       "Difficulty Level         5\n",
+       "Course Rating           31\n",
+       "Course URL            3424\n",
+       "Course Description    3397\n",
+       "Skills                3424\n",
+       "dtype: int64"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [
-    "# Display the number of unique values for each column (useful to identify categorical variables)\n",
-    "print(data.nunique())"
+    "data.nunique()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
-   "id": "6933f893",
+   "execution_count": 8,
+   "id": "3b12b05c",
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "(3424, 7)"
+       "np.int64(98)"
       ]
      },
-     "execution_count": 11,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "# Remove duplicates based on specific columns\n",
-    "data = data.drop_duplicates(subset=['Course Name', 'University', 'Difficulty Level', 'Course Rating',\n",
-    "       'Course URL', 'Course Description'])\n",
-    "data.shape"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "id": "737a0b4d",
-   "metadata": {},
-   "source": [
-    "<h2 style=\"color: #232ED1;\">Check the Distribution of Key Columns</h2>"
+    "data.duplicated().sum()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
-   "id": "4cb5938e",
-   "metadata": {
-    "scrolled": false
-   },
+   "execution_count": 9,
+   "id": "6933f893",
+   "metadata": {},
    "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/var/folders/fh/kyml8tjs61j6z54n3h59ptwh0000gn/T/ipykernel_11894/2219308396.py:14: FutureWarning: \n",
-      "\n",
-      "Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.\n",
-      "\n",
-      "  sns.barplot(x=value_counts.index[:top_n], y=value_counts.values[:top_n], palette='bright')\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1000x600 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/var/folders/fh/kyml8tjs61j6z54n3h59ptwh0000gn/T/ipykernel_11894/2219308396.py:14: FutureWarning: \n",
-      "\n",
-      "Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.\n",
-      "\n",
-      "  sns.barplot(x=value_counts.index[:top_n], y=value_counts.values[:top_n], palette='bright')\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1000x600 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/var/folders/fh/kyml8tjs61j6z54n3h59ptwh0000gn/T/ipykernel_11894/2219308396.py:14: FutureWarning: \n",
-      "\n",
-      "Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.\n",
-      "\n",
-      "  sns.barplot(x=value_counts.index[:top_n], y=value_counts.values[:top_n], palette='bright')\n"
-     ]
-    },
     {
      "data": {
-      "image/png": "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",
       "text/plain": [
-       "<Figure size 1000x600 with 1 Axes>"
+       "(3424, 7)"
       ]
      },
+     "execution_count": 9,
      "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "Total Unique Universities: 184\n"
-     ]
+     "output_type": "execute_result"
     }
    ],
    "source": [
-    "# Function to display and visualize distributions for categorical columns\n",
-    "def display_distribution(column_name, data, top_n=10, sort_by_value=True, visualize=True):\n",
-    "    \n",
-    "    # Get the value counts\n",
-    "    value_counts = data[column_name].value_counts()\n",
-    "\n",
-    "    # Sort values if necessary\n",
-    "    if sort_by_value:\n",
-    "        value_counts = value_counts.sort_values(ascending=False)\n",
-    "    \n",
-    "    # Optional: Plot the distribution\n",
-    "    if visualize:\n",
-    "        plt.figure(figsize=(10, 6))\n",
-    "        sns.barplot(x=value_counts.index[:top_n], y=value_counts.values[:top_n], palette='bright')\n",
-    "        plt.title(f\"Top {top_n} {column_name} Distribution\")\n",
-    "        plt.xlabel(column_name)\n",
-    "        plt.ylabel('Frequency')\n",
-    "        plt.xticks(rotation=45, ha=\"right\")\n",
-    "        plt.tight_layout()\n",
-    "        plt.show()\n",
-    "\n",
-    "# Displaying the distributions\n",
-    "display_distribution('Difficulty Level', data, top_n=5)  # Adjust top_n as needed\n",
-    "display_distribution('Course Rating', data, top_n=5)\n",
-    "display_distribution('University', data, top_n=10, sort_by_value=False)  # For universities, you might not want to sort by count\n",
-    "\n",
-    "# Showing the count of unique universities\n",
-    "print(f\"\\nTotal Unique Universities: {data['University'].nunique()}\")\n"
+    "# Remove duplicates based on specific columns\n",
+    "data = data.drop_duplicates(subset=['Course Name', 'University', 'Difficulty Level', 'Course Rating',\n",
+    "       'Course URL', 'Course Description'])\n",
+    "data.shape"
    ]
   },
   {
@@ -545,40 +387,20 @@
    "id": "b84c5ac7",
    "metadata": {},
    "source": [
-    "<h2 style=\"color: #232ED1;\">Text Preprocessing</h2>"
+    "<h2 style=\"color: #232ED1;\">Text Preprocessing on Training Data</h2>"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 10,
    "id": "e98bdce3",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "                                         Course Name  \\\n",
-      "0  write a feature length screenplay for film or ...   \n",
-      "1  business strategy business model canvas analys...   \n",
-      "2                       silicon thin film solar cell   \n",
-      "3                                finance for manager   \n",
-      "4          retrieve data using singletable sql query   \n",
-      "\n",
-      "                                                tags  \n",
-      "0  write a feature length screenplay for film or ...  \n",
-      "1  business strategy business model canvas analys...  \n",
-      "2  silicon thin film solar cell this course consi...  \n",
-      "3  finance for manager when it come to number the...  \n",
-      "4  retrieve data using singletable sql query in t...  \n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "lemmatizer = WordNetLemmatizer()\n",
     "\n",
     "# Function for text cleaning (removing special characters, stopwords, and lemmatization)\n",
-    "def clean_text(text):\n",
+    "def clean_for_tags(text):\n",
     "    text = re.sub(r'��+', '', text)  # This removes \"��\" or any repeated \"��\" characters\n",
     "    text = re.sub(r'[^\\x00-\\x7F]+', '', text)  # Removes non-ASCII characters\n",
     "    text = re.sub(r'[^a-zA-Z\\s]', '', text)  # Remove anything that is not a letter or space\n",
@@ -586,40 +408,103 @@
     "    text = ' '.join([lemmatizer.lemmatize(word) for word in text.split()])  # Lemmatization\n",
     "    return text\n",
     "\n",
-    "# Apply text cleaning to necessary columns\n",
-    "data['Course Name'] = data['Course Name'].apply(clean_text)\n",
-    "data['Course Description'] = data['Course Description'].apply(clean_text)\n",
-    "data['Skills'] = data['Skills'].apply(clean_text)\n",
+    "training_data = data.copy()\n",
+    "\n",
+    "# Apply clean_for_tags on columns to be used in tags column\n",
+    "training_data['Course Name'] = training_data['Course Name'].apply(clean_for_tags)\n",
+    "training_data['Course Description'] = training_data['Course Description'].apply(clean_for_tags)\n",
+    "training_data['Skills'] = training_data['Skills'].apply(clean_for_tags)\n",
     "\n",
     "# Combine 'Course Name', 'Course Description', and 'Skills' into 'tags'\n",
-    "data['tags'] = data['Course Name'] + ' ' + data['Course Description'] + ' ' + data['Skills']\n",
-    "print(data[['Course Name', 'tags']].head())\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 14,
-   "id": "f6fa97b4",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# Memory efficient datatypes"
+    "data['tags'] = training_data['Course Name'] + ' ' + training_data['Course Description'] + ' ' + training_data['Skills']\n",
+    "\n",
+    "training_data = data[['Course Name', 'tags']]"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
-   "id": "6ce32db5",
+   "execution_count": 11,
+   "id": "d6885a74",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Course Name</th>\n",
+       "      <th>tags</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Write A Feature Length Screenplay For Film Or ...</td>\n",
+       "      <td>write a feature length screenplay for film or ...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Business Strategy: Business Model Canvas Analy...</td>\n",
+       "      <td>business strategy business model canvas analys...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Silicon Thin Film Solar Cells</td>\n",
+       "      <td>silicon thin film solar cell this course consi...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Finance for Managers</td>\n",
+       "      <td>finance for manager when it come to number the...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Retrieve Data using Single-Table SQL Queries</td>\n",
+       "      <td>retrieve data using singletable sql query in t...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                         Course Name  \\\n",
+       "0  Write A Feature Length Screenplay For Film Or ...   \n",
+       "1  Business Strategy: Business Model Canvas Analy...   \n",
+       "2                      Silicon Thin Film Solar Cells   \n",
+       "3                               Finance for Managers   \n",
+       "4       Retrieve Data using Single-Table SQL Queries   \n",
+       "\n",
+       "                                                tags  \n",
+       "0  write a feature length screenplay for film or ...  \n",
+       "1  business strategy business model canvas analys...  \n",
+       "2  silicon thin film solar cell this course consi...  \n",
+       "3  finance for manager when it come to number the...  \n",
+       "4  retrieve data using singletable sql query in t...  "
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
    "source": [
-    "# Convert 'University' and 'Difficulty Level' to categorical to save memory\n",
-    "data['University'] = data['University'].astype('category')\n",
-    "data['Difficulty Level'] = data['Difficulty Level'].astype('category')\n",
-    "\n",
-    "# Optimize 'Course Rating' column to save memory\n",
-    "data['Course Rating'] = data['Course Rating'].apply(lambda x: 0.0 if x == 'Not Calibrated' else pd.to_numeric(x, errors='coerce'))\n",
-    "data['Course Rating'] = data['Course Rating'].astype('float32')\n"
+    "training_data.head()"
    ]
   },
   {
@@ -632,7 +517,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 12,
    "id": "aab37f89",
    "metadata": {},
    "outputs": [
@@ -646,13 +531,21 @@
    ],
    "source": [
     "vectorizer = TfidfVectorizer(stop_words='english', max_features=5000)\n",
-    "tfidf_matrix = vectorizer.fit_transform(data['tags'])\n",
+    "tfidf_matrix = vectorizer.fit_transform(training_data['tags'])\n",
     "print(\"TF-IDF matrix shape:\", tfidf_matrix.shape)"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "42cdb85d",
+   "metadata": {},
+   "source": [
+    "<h2 style=\"color: #232ED1;\">Apply SVD on TF-IDF</h2>"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 13,
    "id": "493e2f7d",
    "metadata": {},
    "outputs": [
@@ -665,10 +558,7 @@
     }
    ],
    "source": [
-    "from sklearn.decomposition import TruncatedSVD\n",
-    "\n",
-    "# Apply SVD to reduce dimensions of the TF-IDF matrix\n",
-    "n_components = 100\n",
+    "n_components = 100 # Reduce to 100 dimensions\n",
     "svd = TruncatedSVD(n_components=n_components, random_state=42)\n",
     "tfidf_matrix = svd.fit_transform(tfidf_matrix)\n",
     "\n",
@@ -685,7 +575,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 14,
    "id": "ab0769d7",
    "metadata": {
     "scrolled": true
@@ -709,79 +599,99 @@
    "id": "62948f78",
    "metadata": {},
    "source": [
-    "<h2 style=\"color: #232ED1;\">Recommendation Function</h2>"
+    "<h2 style=\"color: #232ED1;\">Functions for Recommendation</h2>"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
-   "id": "e79b0a64",
+   "execution_count": 15,
+   "id": "9333fc14",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Recommendations based on 'Finance for Managers':\n",
-      "\n",
-      "- introduction to finance the basic (Adjusted Similarity Score: 0.97, Rating: 4.599999904632568)\n",
-      "- finance for nonfinancial professional (Adjusted Similarity Score: 0.97, Rating: 4.5)\n",
-      "- the language and tool of financial analysis (Adjusted Similarity Score: 0.96, Rating: 4.599999904632568)\n",
-      "- finance for nonfinancial manager (Adjusted Similarity Score: 0.95, Rating: 4.199999809265137)\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
-    "# Ensure 'Course Rating' is numeric, and handle errors gracefully\n",
-    "data['Course Rating'] = pd.to_numeric(data['Course Rating'], errors='coerce')\n",
-    "# Normalize ratings (if ratings are between 1 and 5)\n",
-    "data['Normalized Rating'] = (data['Course Rating'] - data['Course Rating'].min()) / (data['Course Rating'].max() - data['Course Rating'].min())\n",
-    "\n",
-    "def recommend(course_name, top_n=5, threshold=70, min_rating=3):\n",
+    "def normalize_rating(rating_str):\n",
     "    \"\"\"\n",
-    "    Recommend courses based on course name, taking into account both cosine similarity and ratings.\n",
-    "\n",
-    "    Args:\n",
-    "    - course_name (str): Name of the course to base recommendations on.\n",
-    "    - top_n (int): Number of top recommendations to display.\n",
-    "    - threshold (int): Minimum similarity score for a course to be recommended.\n",
-    "    - min_rating (int): Minimum rating a course must have to be recommended.\n",
+    "    Normalize the course rating to a 0-1 scale.\n",
     "    \"\"\"\n",
-    "        \n",
-    "        similarity_scores = list(enumerate(similarity_matrix[course_idx]))\n",
-    "        sorted_courses = sorted(similarity_scores, key=lambda x: x[1], reverse=True)[1:]\n",
-    "        \n",
-    "        recommended_courses = []\n",
-    "        \n",
-    "        # Adjust recommendations using ratings\n",
-    "        for idx, sim_score in sorted_courses[:top_n]:\n",
-    "            course_rating = data.iloc[idx]['Course Rating']\n",
-    "            if pd.isna(course_rating):\n",
-    "                course_rating = data['Course Rating'].mean()  # Fallback to mean rating\n",
-    "\n",
-    "            # Normalize similarity score\n",
-    "            sim_score = float(sim_score)\n",
-    "            sim_min, sim_max = min(sim_score for _, sim_score in sorted_courses), max(sim_score for _, sim_score in sorted_courses)\n",
-    "            normalized_sim_score = (sim_score - sim_min) / (sim_max - sim_min)\n",
+    "    try:\n",
+    "        return (float(rating_str) - 0) / (5 - 0)  # Normalize to 0-1\n",
+    "    except ValueError:\n",
+    "        return 0  "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "id": "a990f12f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def get_recommendations(course_name, data, similarity_matrix, top_n=3, threshold=90, rating_weight=0.05):\n",
+    "    \"\"\"\n",
+    "    Get top N course recommendations based on similarity to the given course name.\n",
+    "    \"\"\"\n",
+    "    course_name = data[data['Course Name'] == course_name]  # Filter data for selected course\n",
+    "    course_idx = course_name.index[0]  # Get the index of the selected course\n",
+    "    similarity_scores = list(enumerate(similarity_matrix[course_idx]))  # Get similarity scores for all courses\n",
+    "    \n",
+    "    recommendations = []\n",
+    "    for idx, similarity_score in sorted(similarity_scores, key=lambda x: x[1], reverse=True)[:top_n]:\n",
+    "        course_data = data.iloc[idx]  # Get course data for the current recommendation\n",
+    "        normalized_rating = normalize_rating(course_data.get('Course Rating', '0'))  # Normalize rating\n",
     "\n",
-    "            # Weighted scoring: 70% similarity, 30% rating\n",
-    "            w_sim = 0.7\n",
-    "            w_rating = 0.3\n",
-    "            adjusted_score = (w_sim * normalized_sim_score) + (w_rating * (course_rating / 5))  # Rating normalized to [0,1]\n",
-    "            \n",
-    "            if course_rating >= min_rating:\n",
-    "                recommended_courses.append((data.iloc[idx]['Course Name'], adjusted_score, course_rating))\n",
-    "        \n",
-    "        recommended_courses = sorted(recommended_courses, key=lambda x: x[1], reverse=True)\n",
-    "        \n",
-    "        # Display recommendations\n",
-    "        for course_name, final_score, rating in recommended_courses:\n",
-    "            print(f\"- {course_name} (Adjusted Similarity Score: {final_score:.2f}, Rating: {rating})\")\n",
-    "    else:\n",
-    "        print(f\"Course '{course_name}' not found.\")\n",
+    "        # Prepare recommendation dictionary with relevant course information\n",
+    "        recommendations.append({\n",
+    "            \"course_name\": course_data['Course Name'],\n",
+    "            \"course_url\": course_data.get('Course URL', ''),\n",
+    "            \"rating\": course_data['Course Rating'],\n",
+    "            \"institution\": course_data.get('University', 'Unknown'),\n",
+    "            \"difficulty_level\": course_data.get('Difficulty Level', 'Unknown'),\n",
+    "            \"similarity\": similarity_score,\n",
+    "            \"final_score\": similarity_score * (1 - rating_weight) + normalized_rating * rating_weight \n",
+    "        })\n",
     "\n",
-    "# Example usage with a course name\n",
-    "recommend('Finance for Managers')\n"
+    "    return sorted(recommendations, key=lambda x: x['final_score'], reverse=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "id": "201f8d31",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[{'course_name': 'Finance for Managers',\n",
+       "  'course_url': 'https://www.coursera.org/learn/operational-finance',\n",
+       "  'rating': '4.8',\n",
+       "  'institution': 'IESE Business School',\n",
+       "  'difficulty_level': 'Intermediate',\n",
+       "  'similarity': np.float64(1.0),\n",
+       "  'final_score': np.float64(0.998)},\n",
+       " {'course_name': 'Finance for Non-Financial Professionals',\n",
+       "  'course_url': 'https://www.coursera.org/learn/finance-for-non-finance-managers',\n",
+       "  'rating': '4.5',\n",
+       "  'institution': 'University of California, Irvine',\n",
+       "  'difficulty_level': 'Conversant',\n",
+       "  'similarity': np.float64(0.832940692838965),\n",
+       "  'final_score': np.float64(0.8362936581970167)},\n",
+       " {'course_name': 'Finance for Non-Financial Managers',\n",
+       "  'course_url': 'https://www.coursera.org/learn/finance-for-non-financial-managers',\n",
+       "  'rating': '4.2',\n",
+       "  'institution': 'Emory University',\n",
+       "  'difficulty_level': 'Beginner',\n",
+       "  'similarity': np.float64(0.8315387472926317),\n",
+       "  'final_score': np.float64(0.8319618099280001)}]"
+      ]
+     },
+     "execution_count": 18,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "get_recommendations('Finance for Managers', data, similarity_matrix)"
    ]
   },
   {
@@ -794,18 +704,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 19,
    "id": "8c3a275a",
    "metadata": {},
    "outputs": [],
    "source": [
-    "# Saving the trained model\n",
-    "pickle.dump(similarity_matrix, open('similarity_matrix.pkl', 'wb'))\n",
-    "pickle.dump(vectorizer, open('tfidf_vectorizer.pkl', 'wb'))\n",
-    "\n",
-    "# In a web application or future use, you can load the model:\n",
-    "# similarity_matrix = pickle.load(open('similarity_matrix.pkl', 'rb'))\n",
-    "# vectorizer = pickle.load(open('tfidf_vectorizer.pkl', 'rb'))\n"
+    "pickle.dump(similarity_matrix, open('similarity_matrix.pkl', 'wb'))"
    ]
   }
  ],