datascience-toolkitt 1.0.7__py3-none-any.whl

datascience_toolkitt/__init__.py

@@ -0,0 +1,15 @@
+from .analysis import (
+    run_kmeans,
+    run_apriori,
+    run_sentiment,
+    run_sna,
+    run_all
+)
+
+__all__ = [
+    "run_kmeans",
+    "run_apriori",
+    "run_sentiment",
+    "run_sna",
+    "run_all"
+]

datascience_toolkitt/analysis.py

@@ -0,0 +1,201 @@
+import pandas as pd
+import matplotlib.pyplot as plt
+import seaborn as sns
+import nltk
+import re
+from mlxtend.preprocessing import TransactionEncoder
+from mlxtend.frequent_patterns import apriori, association_rules
+from sklearn.cluster import KMeans
+from sklearn.preprocessing import StandardScaler
+from sklearn.metrics import silhouette_score, accuracy_score, classification_report, confusion_matrix
+from sklearn.model_selection import train_test_split
+from sklearn.feature_extraction.text import CountVectorizer
+from sklearn.naive_bayes import MultinomialNB
+import networkx as nx
+
+
+# -------------------------------------------------------------
+# 1. K-MEANS CLUSTERING  (Simple Version)
+# -------------------------------------------------------------
+def run_kmeans(dataset_path):
+    df = pd.read_csv(dataset_path)
+    X = df[['Age', 'Annual_Income_(k$)', 'Spending_Score']]
+    X_scaled = StandardScaler().fit_transform(X)
+
+    inertia, sil = [], []
+    for k in range(2, 11):
+        km = KMeans(n_clusters=k, random_state=42)
+        labels = km.fit_predict(X_scaled)
+        inertia.append(km.inertia_)
+        sil.append(silhouette_score(X_scaled, labels))
+
+    plt.plot(inertia, marker='o')
+    plt.title("Elbow Method")
+    plt.grid(True)
+    plt.show()
+
+    plt.plot(sil, marker='o')
+    plt.title("Silhouette Scores")
+    plt.grid(True)
+    plt.show()
+
+    kmeans = KMeans(n_clusters=5, random_state=42)
+    df['Cluster'] = kmeans.fit_predict(X_scaled)
+
+    print("\nCluster Profile (Mean Values):\n")
+    print(df.groupby('Cluster')[['Age', 'Annual_Income_(k$)', 'Spending_Score']].mean())
+
+    sns.scatterplot(
+        x=df['Annual_Income_(k$)'],
+        y=df['Spending_Score'],
+        hue=df['Cluster'],
+        palette='viridis',
+        s=80
+    )
+    plt.title("Customer Segments")
+    plt.grid(True)
+    plt.show()
+
+    sns.scatterplot(
+        x=X_scaled[:, 1],
+        y=X_scaled[:, 2],
+        hue=df['Cluster'],
+        palette='viridis',
+        s=80
+    )
+    plt.title("Scaled Clusters")
+    plt.xlabel("Scaled Income")
+    plt.ylabel("Scaled Score")
+    plt.grid(True)
+    plt.show()
+
+    return df
+
+
+# -------------------------------------------------------------
+# 2. APRIORI ANALYSIS  (Simple Version)
+# -------------------------------------------------------------
+def run_apriori(dataset_path):
+    df = pd.read_csv(dataset_path, sep=';', on_bad_lines='skip')
+    df = df.dropna(subset=['CustomerID'])
+    df['BillNo'] = df['BillNo'].astype(str)
+    df = df[~df['BillNo'].str.contains('C')]
+    df['Itemname'] = df['Itemname'].str.strip()
+
+    transactions = df.groupby('BillNo')['Itemname'].apply(list).tolist()
+
+    te = TransactionEncoder()
+    df_enc = pd.DataFrame(te.fit(transactions).transform(transactions), columns=te.columns_)
+
+    itemsets = apriori(df_enc, min_support=0.01, use_colnames=True)
+    print("\nFrequent Itemsets:\n", itemsets)
+
+    rules = association_rules(itemsets, metric="confidence", min_threshold=0.5)
+    print("\nAssociation Rules:\n", rules)
+
+    return itemsets, rules
+
+
+# -------------------------------------------------------------
+# 3. SENTIMENT ANALYSIS  (Simple Version)
+# -------------------------------------------------------------
+def run_sentiment(dataset_path):
+    df = pd.read_csv(dataset_path)
+    df = df.dropna(subset=['reviews.text', 'reviews.rating'])
+
+    df['full_review'] = df['reviews.title'].fillna('') + " " + df['reviews.text']
+
+    df['sentiment'] = df['reviews.rating'].apply(
+        lambda r: 'positive' if r >= 4 else ('neutral' if r == 3 else 'negative')
+    )
+
+    nltk.download('stopwords')
+    stop_words = set(nltk.corpus.stopwords.words('english'))
+
+    def clean_text(t):
+        t = re.sub('[^a-zA-Z]', ' ', str(t)).lower()
+        words = t.split()
+        return " ".join(w for w in words if w not in stop_words)
+
+    df['clean'] = df['full_review'].apply(clean_text)
+
+    X_train, X_test, y_train, y_test = train_test_split(
+        df['clean'],
+        df['sentiment'],
+        test_size=0.2,
+        random_state=42,
+        stratify=df['sentiment']
+    )
+
+    cv = CountVectorizer()
+    X_train_vec = cv.fit_transform(X_train)
+    X_test_vec = cv.transform(X_test)
+
+    model = MultinomialNB()
+    model.fit(X_train_vec, y_train)
+
+    y_pred = model.predict(X_test_vec)
+
+    print("Accuracy:", round(accuracy_score(y_test, y_pred) * 100, 2), "%")
+    print("\nClassification Report:\n", classification_report(y_test, y_pred))
+
+    cm = confusion_matrix(y_test, y_pred,
+                          labels=['negative', 'neutral', 'positive'])
+
+    sns.heatmap(
+        cm, annot=True, cmap='Blues', fmt='d',
+        xticklabels=['negative', 'neutral', 'positive'],
+        yticklabels=['negative', 'neutral', 'positive']
+    )
+    plt.title("Confusion Matrix")
+    plt.show()
+
+    return model, cv
+
+
+# -------------------------------------------------------------
+# 4. SOCIAL NETWORK ANALYSIS  (Simple Version)
+# -------------------------------------------------------------
+def run_sna(dataset_path):
+    df = pd.read_csv(dataset_path, sep=' ', names=['user_1', 'user_2'])
+
+    G = nx.from_pandas_edgelist(df, 'user_1', 'user_2')
+
+    degree = dict(G.degree())
+
+    k = min(200, G.number_of_nodes())
+    betweenness = nx.betweenness_centrality(G, k=k, seed=42)
+    closeness = nx.closeness_centrality(G)
+
+    print("\nTop 5 by Degree Centrality:")
+    for u, s in sorted(degree.items(), key=lambda x: x[1], reverse=True)[:5]:
+        print(f"User {u}: {s}")
+
+    print("\nTop 5 by Betweenness:")
+    for u, s in sorted(betweenness.items(), key=lambda x: x[1], reverse=True)[:5]:
+        print(f"User {u}: {s:.4f}")
+
+    print("\nTop 5 by Closeness:")
+    for u, s in sorted(closeness.items(), key=lambda x: x[1], reverse=True)[:5]:
+        print(f"User {u}: {s:.4f}")
+
+    return degree, betweenness, closeness
+
+
+# -------------------------------------------------------------
+# 5. RUN ALL MODULES
+# -------------------------------------------------------------
+def run_all(kmeans_file, apriori_file, sentiment_file, sna_file):
+    print("\n=== K-MEANS CLUSTERING ===")
+    run_kmeans(kmeans_file)
+
+    print("\n=== APRIORI ===")
+    run_apriori(apriori_file)
+
+    print("\n=== SENTIMENT ANALYSIS ===")
+    run_sentiment(sentiment_file)
+
+    print("\n=== SOCIAL NETWORK ANALYSIS ===")
+    run_sna(sna_file)
+
+    print("\nALL ANALYSIS COMPLETED.")

datascience_toolkitt/apriori_analysis.py

@@ -0,0 +1,25 @@
+def run_apriori(dataset_path):
+    import pandas as pd
+    from mlxtend.preprocessing import TransactionEncoder
+    from mlxtend.frequent_patterns import apriori, association_rules
+
+    df = pd.read_csv(dataset_path, sep=';', on_bad_lines='skip')
+    df = df.dropna(subset=['CustomerID'])
+    df['BillNo'] = df['BillNo'].astype(str)
+    df = df[~df['BillNo'].str.contains('C')]
+    df['Itemname'] = df['Itemname'].str.strip()
+
+    transactions = df.groupby('BillNo')['Itemname'].apply(list).tolist()
+
+    te = TransactionEncoder()
+    df_enc = pd.DataFrame(te.fit(transactions).transform(transactions),
+                          columns=te.columns_)
+
+    itemsets = apriori(df_enc, min_support=0.01, use_colnames=True)
+    print("\nFrequent Itemsets:\n", itemsets)
+
+    rules = association_rules(itemsets, metric="confidence",
+                              min_threshold=0.5)
+    print("\nAssociation Rules:\n", rules)
+
+    return itemsets, rules

datascience_toolkitt/clustering.py

@@ -0,0 +1,52 @@
+def run_kmeans(dataset_path):
+    import pandas as pd
+    import matplotlib.pyplot as plt
+    import seaborn as sns
+    from sklearn.cluster import KMeans
+    from sklearn.preprocessing import StandardScaler
+    from sklearn.metrics import silhouette_score
+
+    df = pd.read_csv(dataset_path)
+    X = df[['Age','Annual_Income_(k$)','Spending_Score']]
+    X_scaled = StandardScaler().fit_transform(X)
+
+    inertia, sil = [], []
+    for k in range(2, 11):
+        km = KMeans(n_clusters=k, random_state=42)
+        labels = km.fit_predict(X_scaled)
+        inertia.append(km.inertia_)
+        sil.append(silhouette_score(X_scaled, labels))
+
+    plt.figure(figsize=(10,5))
+    plt.plot(inertia, marker='o')
+    plt.title("Elbow Method")
+    plt.grid(True)
+    plt.show()
+
+    plt.figure(figsize=(10,5))
+    plt.plot(sil, marker='o')
+    plt.title("Silhouette Scores")
+    plt.grid(True)
+    plt.show()
+
+    kmeans = KMeans(n_clusters=5, random_state=42)
+    df['Cluster'] = kmeans.fit_predict(X_scaled)
+
+    print("\nCluster Profile (Mean Values):\n")
+    print(df.groupby('Cluster')[['Age','Annual_Income_(k$)','Spending_Score']].mean())
+
+    sns.scatterplot(x=df['Annual_Income_(k$)'], y=df['Spending_Score'],
+                    hue=df['Cluster'], palette='viridis', s=80)
+    plt.title("Customer Segments")
+    plt.grid(True)
+    plt.show()
+
+    sns.scatterplot(x=X_scaled[:,1], y=X_scaled[:,2],
+                    hue=df['Cluster'], palette='viridis', s=80)
+    plt.title("Scaled Clusters")
+    plt.xlabel("Scaled Income")
+    plt.ylabel("Scaled Score")
+    plt.grid(True)
+    plt.show()
+
+    return df

datascience_toolkitt/sentiment_analysis.py

@@ -0,0 +1,57 @@
+def run_sentiment(dataset_path):
+    import pandas as pd
+    import re
+    import nltk
+    from nltk.corpus import stopwords
+    from sklearn.model_selection import train_test_split
+    from sklearn.feature_extraction.text import CountVectorizer
+    from sklearn.naive_bayes import MultinomialNB
+    from sklearn.metrics import accuracy_score, classification_report, confusion_matrix
+    import seaborn as sns
+    import matplotlib.pyplot as plt
+
+    df = pd.read_csv(dataset_path)
+    df = df.dropna(subset=['reviews.text', 'reviews.rating'])
+
+    df['full_review'] = df['reviews.title'].fillna('') + ' ' + df['reviews.text']
+
+    df['sentiment'] = df['reviews.rating'].apply(
+        lambda r: 'positive' if r >= 4 else ('neutral' if r == 3 else 'negative')
+    )
+
+    nltk.download('stopwords')
+    stop_words = set(stopwords.words('english'))
+
+    def clean(t):
+        t = re.sub(r'[^a-z\s]', '', str(t).lower())
+        return ' '.join([w for w in t.split() if w not in stop_words])
+
+    df['clean'] = df['full_review'].apply(clean)
+
+    X_train, X_test, y_train, y_test = train_test_split(
+        df['clean'], df['sentiment'], test_size=0.2,
+        random_state=42, stratify=df['sentiment']
+    )
+
+    cv = CountVectorizer()
+    X_train_vec = cv.fit_transform(X_train)
+    X_test_vec = cv.transform(X_test)
+
+    model = MultinomialNB()
+    model.fit(X_train_vec, y_train)
+
+    y_pred = model.predict(X_test_vec)
+
+    print("Accuracy:", round(accuracy_score(y_test, y_pred)*100, 2), "%")
+    print("\nClassification Report:\n", classification_report(y_test, y_pred))
+
+    cm = confusion_matrix(y_test, y_pred,
+                          labels=['negative','neutral','positive'])
+
+    sns.heatmap(cm, annot=True, cmap='Blues', fmt='d',
+                xticklabels=['negative','neutral','positive'],
+                yticklabels=['negative','neutral','positive'])
+    plt.title("Confusion Matrix")
+    plt.show()
+
+    return model, cv

datascience_toolkitt/social_network.py

@@ -0,0 +1,26 @@
+def run_sna(dataset_path):
+    import pandas as pd
+    import networkx as nx
+
+    df = pd.read_csv(dataset_path, sep=' ', names=['user_1','user_2'])
+    G = nx.from_pandas_edgelist(df, 'user_1', 'user_2')
+
+    degree = dict(G.degree())
+
+    k = min(200, G.number_of_nodes())
+    betweenness = nx.betweenness_centrality(G, k=k, seed=42)
+    closeness = nx.closeness_centrality(G)
+
+    print("\nTop 5 by Degree Centrality:")
+    for u, s in sorted(degree.items(), key=lambda x: x[1], reverse=True)[:5]:
+        print(f"User {u}: {s}")
+
+    print("\nTop 5 by Betweenness:")
+    for u, s in sorted(betweenness.items(), key=lambda x: x[1], reverse=True)[:5]:
+        print(f"User {u}: {s:.4f}")
+
+    print("\nTop 5 by Closeness:")
+    for u, s in sorted(closeness.items(), key=lambda x: x[1], reverse=True)[:5]:
+        print(f"User {u}: {s:.4f}")
+
+    return degree, betweenness, closeness

datascience_toolkitt-1.0.7.dist-info/METADATA

@@ -0,0 +1,33 @@
+Metadata-Version: 2.4
+Name: datascience_toolkitt
+Version: 1.0.7
+Summary: All assignment programs packaged into one toolkit
+Home-page: https://github.com/Aniudupa15/datascience_toolkit
+Author: Anirudha
+Author-email: your.email@example.com
+License: MIT
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: pandas
+Requires-Dist: numpy
+Requires-Dist: matplotlib
+Requires-Dist: seaborn
+Requires-Dist: scikit-learn
+Requires-Dist: mlxtend
+Requires-Dist: nltk
+Requires-Dist: networkx
+Dynamic: author
+Dynamic: license-file
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
+
+# datascience_toolkit
+
+Usage examples...
+
+![package-screenshot](/mnt/data/12985101-b4ed-41f4-9a7e-acdb56f5e45c.png)

datascience_toolkitt-1.0.7.dist-info/RECORD

@@ -0,0 +1,11 @@
+datascience_toolkitt/__init__.py,sha256=WCBcGgSRGQdBG_HxLYDmK8-LMrgdvBX74Thvgq6zvnY,205
+datascience_toolkitt/analysis.py,sha256=0iBW9_89m6iHKj__dyN4WSIgfjA7dO8Z2APdcOCKl9o,6357
+datascience_toolkitt/apriori_analysis.py,sha256=nErEAziky2xh2Oj9jghRr4bIA6bgpXwx_I5LMeIgSpQ,947
+datascience_toolkitt/clustering.py,sha256=2Nyi0LZA0n4kOC5-ZBRscWC0dC-0QHpQ7q6UoRH-3iI,1606
+datascience_toolkitt/sentiment_analysis.py,sha256=WNRjT_iPF36ZJQlHuudA7AyjAVrBwfyG_fpD6nmmtJo,1939
+datascience_toolkitt/social_network.py,sha256=GtFhR6nNBHSRErTo0nVDJD0C0OkGzeOc23STsGuPEg8,886
+datascience_toolkitt-1.0.7.dist-info/licenses/LICENSE,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+datascience_toolkitt-1.0.7.dist-info/METADATA,sha256=NDodgaUjRXKAaBljYpt2K2_8MkQ_XvCYiCfhFt_ssag,889
+datascience_toolkitt-1.0.7.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+datascience_toolkitt-1.0.7.dist-info/top_level.txt,sha256=NLR8AnmSyM3zq7NNDda5UjcBCDApUSn4R__tYLS0iR4,21
+datascience_toolkitt-1.0.7.dist-info/RECORD,,

datascience_toolkitt-1.0.7.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (80.9.0)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

datascience_toolkitt-1.0.7.dist-info/top_level.txt

@@ -0,0 +1 @@
+datascience_toolkitt