Update pages/6_Feature_Engineering.py

pages/6_Feature_Engineering.py

@@ -565,5 +565,88 @@ elif file_type == "Term Frequency - Inverse Document Frequency(TF-IDF)":
     unsafe_allow_html=True,
     )
 
-st.markdown("<p style='text-align: center; font-size: 18px;'><strong>TF-IDF effectively balances word significance and document relevance! 🚀</strong></p>", unsafe_allow_html=True)
-    
+    st.markdown("<h1 class='title'>📌 TF-IDF Key Insights</h1>", unsafe_allow_html=True)
+
+    st.markdown(
+    """
+    <div class='box'>
+        <h3 style='color: #6A0572;'>📈 Case 1: High TF-IDF Values</h3>
+        <ul>
+            <li>If the word appears <strong>frequently</strong> in a document → <span class='highlight'>High TF-IDF</span></li>
+        </ul>
+    </div>
+    """,
+    unsafe_allow_html=True,
+    )
+
+    st.markdown(
+    """
+    <div class='box'>
+        <h3 style='color: #6A0572;'>📉 Case 2: Low TF-IDF Values</h3>
+        <ul>
+            <li>If the word appears <strong>rarely</strong> in a document → <span class='highlight'>Low TF-IDF</span></li>
+            <li>TF is always in the range: <strong>[0 - 1]</strong></li>
+            <li>IDF is in the range: <strong>[0 - ∞)</strong></li>
+        </ul>
+    </div>
+    """,
+    unsafe_allow_html=True,
+    )
+
+    st.markdown(
+    """
+    <div class='box'>
+        <h3 style='color: #6A0572;'>📊 Understanding TF (Term Frequency)</h3>
+        <ul>
+            <li>TF gives <strong>more importance</strong> to words that occur <strong>frequently</strong> in a document.</li>
+            <li>As the word frequency <span class='highlight'>increases</span> → TF <span class='highlight'>increases</span>.</li>
+        </ul>
+    </div>
+    """,
+    unsafe_allow_html=True,
+    )
+
+    st.markdown(
+    """
+    <div class='box'>
+        <h3 style='color: #6A0572;'>📉 Understanding IDF (Inverse Document Frequency)</h3>
+        <ul>
+            <li>IDF Formula: <span class='highlight'>IDF(wᵢ, C) = log(N/n)</span></li>
+            <li><strong>N:</strong> Total number of documents</li>
+            <li><strong>n:</strong> Number of documents containing the word</li>
+        </ul>
+    </div>
+    """,
+    unsafe_allow_html=True,
+    )
+
+    st.markdown(
+    """
+    <div class='formula'>
+        <strong>When n is small:</strong> <br>
+        - N/n increases → log(N/n) increases ⬆️ <br>
+        - Word is rare in the corpus → Higher importance in IDF <br><br>
+        <strong>When n is large:</strong> <br>
+        - N/n decreases → log(N/n) decreases ⬇️ <br>
+        - Word is common → Lower importance in IDF <br><br>
+        <strong>When N = n:</strong> log(N/n) = 0 (word appears in every document)
+    </div>
+    """,
+    unsafe_allow_html=True,
+    )
+
+    st.markdown(
+    """
+    <div class='box'>
+        <h3 style='color: #6A0572;'>📌 TF-IDF Calculation</h3>
+        <ul>
+            <li><strong>TF</strong> focuses on words <strong>frequent</strong> in a document.</li>
+            <li><strong>IDF</strong> focuses on words <strong>rare</strong> in the corpus.</li>
+            <li><span class='highlight'>TF-IDF is high</span> for words that appear <strong>often in a document</strong> but <strong>rarely in the corpus</strong>.</li>
+        </ul>
+    </div>
+    """,
+    unsafe_allow_html=True,
+    )
+
+