vault backup: 2026-01-27 15:17:32

Documents/Arbeit/IFN/Programmieren WiSe 25 26/Notizen.md

@@ -10,7 +10,7 @@ Fehlende Daten:
 - a22ad77be0d478b5fe34d1167d4dbb3a
 - a96f7c881d27b739c38b81da2058a2fb
 - c6679ce22dfd5646f81d40a8dbb0236b
-
+ 
 Beispiele:
 
 >[!danger]
@@ -56,40 +56,25 @@ Ideen:
 
 
 ```python
-# Plot
-fig, ax = plt.subplots(figsize=(8,5))
-ax.set_title(
-"Verteiltung Durchschn. Bruttomonatsverdienste ohne Sonderz. (nach Geschlecht)"
-)
-sb.kdeplot(
-	female, label="Weiblich",
-	color="#fe640b", fill=True,
-	alpha=0.5, ax=ax
-	)
-sb.kdeplot(
-	male, label="Männlich",
-	color="#40a02b", fill=True,
-	alpha=0.5, ax=ax
-	)
-ax.legend()
+# Binäre Variable: Vegan = 1, Nicht-Vegan = 0
+df['ist_vegan'] = (df['Ernährung'] == 'Vegan').astype(int)
 
-text = f'''t‑stat: {t_stat:.2f}
-p‑value: {p_value:.3f}
-Pay Gap: {unbereinigter_gpg:.2f}%
-Abs. Diff: {bruttoeinkommen_diff:.2f}€'''
+# Filter: mindestens 2 Instagram-Accounts UND mindestens einer privat
+h2_data = df[
+	(df['Instagram_Anzahl'] >= 2) & (df['Instagram_Privat'] == 'Ja')
+	]
 
-ax.text(0.95, 0.75, text,
-        transform=ax.transAxes,
-        fontsize=10,
-        verticalalignment='top',
-        horizontalalignment='right',
-        bbox=dict(
-	        facecolor='white',
-	        edgecolor='black',
-	        pad=5, alpha=0.8
-	        )
-	    )
-plt.show()
+# Gruppen
+vegan_mit = h2_data['ist_vegan']
+vegan_ohne = df[
+	(df['Instagram_Anzahl'] < 2) | (df['Instagram_Privat'] != 'Ja')
+	]['ist_vegan']
+
+# Independent t-test
+t_stat, p_value = ttest_ind(vegan_mit, vegan_ohne)
+
+vegan_pct_mit = vegan_mit.mean() * 100
+vegan_pct_ohne = vegan_ohne.mean() * 100
 ```
 
 
@@ -98,5 +83,11 @@ $$\text{Birth Rate} = \frac{B}{P} * 1000$$
 # Projekte
 - [Social Media Addiction](https://www.kaggle.com/datasets/adilshamim8/social-media-addiction-vs-relationships)
 - [Movies & Shows](https://www.kaggle.com/datasets/shivamb/amazon-prime-movies-and-tv-shows)
-- 8
+
+
+
+
+Annemike Rörig anschreiben
+
+