Update app.py

app.py

@@ -2,50 +2,126 @@ import pandas as pd
 import plotly.express as px
 from datasets import load_dataset
 import gradio as gr
+import yfinance as yf
 
 # --- LOAD DATASET ---
 df = pd.DataFrame(load_dataset("SelmaNajih001/NewsSentiment")["train"])
-df=df[(df["Company"]=="Tesla") | (df["Company"]=="Microsoft") | (df["Company"]=="Apple") | (df["Company"]=="Facebook")]
+df = df[df["Company"].isin(["Tesla", "Microsoft", "Apple", "Facebook", "Google"])]
 
 # --- CONVERT DATE TO DATETIME SAFELY ---
 df['Date'] = pd.to_datetime(df['Date'], errors='coerce')
-
 df['Year'] = df['Date'].dt.year
-df['Date'] = pd.to_datetime(df['Date'])
 df['Month'] = df['Date'].dt.to_period('M')
 df['Day'] = df['Date'].dt.date
 
+# --- TICKERS YAHOO FINANCE ---
+TICKERS = {
+    "Tesla": "TSLA",
+    "Microsoft": "MSFT",
+    "Apple": "AAPL",
+    "Facebook": "META",  # ex FB
+    "Google": "GOOGL",
+    "NASDAQ": "^IXIC"
+}
+
+# --- FETCH STOCK PRICES ---
+prices = {}
+for company, ticker in TICKERS.items():
+    start_date = df['Date'].min()  # prendi la data minima del dataset per tutti
+    end_date = pd.Timestamp.today()
+    df_prices = yf.download(ticker, start=start_date, end=end_date)[['Close']].reset_index()
+    df_prices['Date'] = pd.to_datetime(df_prices['Date'])
+    prices[company] = df_prices
+
+# --- MERGE PRICES INTO DATASET ---
+df_merged = df.copy()
+for company, df_price in prices.items():
+    if company == "NASDAQ":
+        continue  # NASDAQ lo useremo solo per confronto aggregato
+    mask = df_merged['Company'] == company
+    df_merged.loc[mask, 'Close'] = pd.merge(
+        df_merged[mask],
+        df_price,
+        on='Date',
+        how='left'
+    )['Close'].values
+
 # --- GRADIO FUNCTION ---
 def show_sentiment(selected_companies, aggregation="Day"):
-    # filtraggio aziende
     if selected_companies:
-        df_filtered = df[df['Company'].isin(selected_companies)].copy()
+        df_filtered = df_merged[df_merged['Company'].isin(selected_companies)].copy()
     else:
-        df_filtered = df.copy()
-    
-    # scelta colonna per aggregazione
+        # Se non viene selezionata alcuna azienda, usa tutte le aziende per sentiment aggregato
+        df_filtered = df_merged.copy()
+        selected_companies = ["NASDAQ"]  # per mostrare anche NASDAQ
+        # Creiamo un dataframe NASDAQ per unire al grafico
+        df_nasdaq = prices["NASDAQ"].copy()
+        df_nasdaq = df_nasdaq.rename(columns={'Close': 'Score'})
+        df_nasdaq['Company'] = 'NASDAQ'
+        df_nasdaq[group_col := 'Date'] = df_nasdaq['Date']
+        df_filtered = pd.concat([df_filtered, df_nasdaq], ignore_index=True, sort=False)
+
+    # Determina colonna di aggregazione
     if aggregation == "Day":
         group_col = "Day"
     elif aggregation == "Month":
         group_col = "Month"
-        # Convertiamo Period in datetime per Plotly
         df_filtered['Month'] = df_filtered['Month'].dt.to_timestamp()
     elif aggregation == "Year":
         group_col = "Year"
     else:
-        group_col = "Day"  # default
+        group_col = "Day"
 
-    # raggruppamento
-    if selected_companies:
-        df_grouped = df_filtered.groupby([group_col, 'Company'])['Score'].sum().reset_index()
-        fig = px.line(df_grouped, x=group_col, y='Score', color='Company',
-                      title=f"Sentiment Score by {aggregation} per Company")
+    # Raggruppamento con sentiment e prezzo
+    if "NASDAQ" in selected_companies:
+        df_grouped = df_filtered.groupby([group_col, 'Company']).agg({
+            'Score': 'sum',
+            'Close': 'last'
+        }).reset_index()
     else:
-        df_grouped = df_filtered.groupby(group_col)['Score'].sum().reset_index()
-        fig = px.line(df_grouped, x=group_col, y='Score',
-                      title=f"General Sentiment Score by {aggregation}")
+        df_grouped = df_filtered.groupby([group_col, 'Company']).agg({
+            'Score': 'sum',
+            'Close': 'last'
+        }).reset_index()
+
+    # --- CREAZIONE FIGURA ---
+    fig = px.line(df_grouped, x=group_col, y='Score', color='Company',
+                  title=f"Sentiment Score by {aggregation} per Company")
+
+    # Aggiungi linea prezzo sul secondary y
+    for company in selected_companies:
+        if company == "NASDAQ":
+            df_c = df_grouped[df_grouped['Company'] == 'NASDAQ']
+            fig.add_scatter(
+                x=df_c[group_col],
+                y=df_c['Score'],
+                mode='lines',
+                name="NASDAQ Index",
+                yaxis="y2",
+                line=dict(dash='dot', color='yellow')
+            )
+        else:
+            df_c = df_grouped[df_grouped['Company'] == company]
+            fig.add_scatter(
+                x=df_c[group_col],
+                y=df_c['Close'],
+                mode='lines',
+                name=f"{company} Price",
+                yaxis="y2",
+                line=dict(dash='dot')
+            )
 
-    return df_grouped.tail(30), fig  # ultime 30 righe per leggibilità
+    fig.update_layout(
+        yaxis_title="Sentiment Score",
+        yaxis2=dict(
+            title="Stock Price / NASDAQ Index",
+            overlaying="y",
+            side="right"
+        ),
+        hovermode="x unified"
+    )
+
+    return df_grouped.tail(30), fig
 
 # --- GRADIO INTERFACE ---
 companies = sorted(df['Company'].unique().tolist())
@@ -67,10 +143,11 @@ demo = gr.Interface(
     ],
     outputs=[
         gr.Dataframe(label="Sentiment Table", type="pandas"),
-        gr.Plot(label="Sentiment Trend"),
+        gr.Plot(label="Sentiment & Stock Price Trend"),
     ],
-    title="Dynamic Sentiment Dashboard",
-    description="Shows sentiment scores aggregated by day, month, or year. Positive = +score, Negative = -score, Neutral = 0."
+    title="Dynamic Sentiment & Stock Price Dashboard",
+    description="Shows sentiment scores aggregated by day, month, or year. Overlay stock prices from Yahoo Finance and NASDAQ index for comparison."
 )
 
 demo.launch()
+