Streamlit_Association/main.py at main · J3ndra/Streamlit_Association · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
import streamlit as st
import pandas as pd
import itertools
import plotly.express as px
from mlxtend.preprocessing import TransactionEncoder
from mlxtend.frequent_patterns import apriori, association_rules, fpgrowth

st.title('Aplikasi Web Data Mining Asosiasi!')

st.write("""
         # Dataset
         ##### Select dataset!
         """)

nama_algoritma = st.sidebar.selectbox(
    'Pilih Algoritma',
    ('Apriori', 'FP Growth'),
)

uploaded_file = st.file_uploader("Choose a CSV file", type="csv")
if uploaded_file is not None:
    data = pd.read_csv(uploaded_file)

    st.write(data)

    st.write(f"##### Data shape {data.shape}")

    st.write("# Preprocessing")
    products_list = [product.split(",") for product in data['products'].values]
    te = TransactionEncoder()
    te_data = te.fit_transform(products_list)
    df_encoded = pd.DataFrame(te_data, columns=te.columns_)
    st.dataframe(df_encoded)

    st.write("##### Sum the data")
    sums = df_encoded.sum().sort_values(ascending=False)
    st.write(sums)

    # Custom Function
    def ar_iterations(data, num_iter=1, support_value=0.1, iterationIndex=None):
        # Next Iterations
        def ar_calculation(iterationIndex=iterationIndex):
            # Calculation of support value
            value = []
            for i in range(0, len(iterationIndex)):
                result = data.T.loc[iterationIndex[i]].sum()
                result = len(
                    result[result == data.T.loc[iterationIndex[i]].shape[0]]) / data.shape[0]
                value.append(result)
            # Bind results
            result = pd.DataFrame(value, columns=["Support"])
            result["index"] = [tuple(i) for i in iterationIndex]
            result['length'] = result['index'].apply(lambda x: len(x))
            result = result.set_index("index").sort_values(
                "Support", ascending=False)
            # Elimination by Support Value
            result = result[result.Support > support_value]
            return result

        # First Iteration
        first = pd.DataFrame(df_encoded.T.sum(
            axis=1) / df_encoded.shape[0], columns=["Support"]).sort_values("Support", ascending=False)
        first = first[first.Support > support_value]
        first["length"] = 1

        if num_iter == 1:
            res = first.copy()
        # Second Iteration
        elif num_iter == 2:
            second = list(itertools.combinations(first.index, 2))
            second = [list(i) for i in second]
            res = ar_calculation(second)
        # All Iterations > 2
        else:
            nth = list(itertools.combinations(
                set(list(itertools.chain(*iterationIndex))), num_iter))
            nth = [list(i) for i in nth]
            res = ar_calculation(nth)

        return res

    st.write("# Iteration")
    st.write("##### First Iteration")
    iteration1 = ar_iterations(df_encoded, num_iter=1, support_value=0.1)
    st.write(iteration1)

    st.write("##### Second Iteration")
    iteration2 = ar_iterations(df_encoded, num_iter=2, support_value=0.1)
    st.write(iteration2)

    st.write("##### Third Iteration")
    iteration3 = ar_iterations(df_encoded, num_iter=3, support_value=0.01,
                               iterationIndex=iteration2.index)
    st.write(iteration3)

    st.write("##### Fourth Iteration")
    iteration4 = ar_iterations(df_encoded, num_iter=4, support_value=0.01,
                               iterationIndex=iteration3.index)
    st.write(iteration4)

    st.write(f"""
             # Algorithm
             ##### {nama_algoritma}
             """)

    if nama_algoritma == 'Apriori':
        freq_items = apriori(df_encoded, min_support=0.1,
                             use_colnames=True, verbose=1)
    else:
        freq_items = fpgrowth(df_encoded, min_support=0.1,
                              use_colnames=True, verbose=1)

    st.write(freq_items.sort_values("support", ascending=False))

    st.write(f"##### {nama_algoritma} Top 5")
    st.dataframe(freq_items.head())

    st.write(f"##### {nama_algoritma} Bottom 5")
    st.dataframe(freq_items.tail())

    st.write("# Association Rules & Info")
    df_ar = association_rules(
        freq_items, metric="confidence", min_threshold=0.5)
    st.write(df_ar)

    st.write("##### Explanation")
    st.write("""
             1. antecedent support : memberitahu kita kemungkinan produk anteseden saja
             2. consequent support : memberitahu kita kemungkinan konsekuensi produk saja
             3. support            : memberitahu kita nilai dari kedua produk (Anteseden dan Konsekuen)
             4. confidence         : memberitahu kita indikasi seberapa sering aturan tersebut ditemukan benar.
             """)

    st.write("# Show table by filtering from Streamlit")
    support_value = st.slider("Select Support Value", 0.0, 1.0, 0.5, 0.05)
    confidence_value = st.slider(
        "Select Confidence Value", 0.0, 1.0, 0.5, 0.05)
    filtered_df = df_ar[(df_ar.support > support_value) &
                        (df_ar.confidence > confidence_value)]
    sorted_df = filtered_df.sort_values("confidence", ascending=False)
    st.write(sorted_df)

    # Create plot
    sorted_df['antecedents'] = sorted_df['antecedents'].apply(
        lambda x: ', '.join(list(x)))
    sorted_df['consequents'] = sorted_df['consequents'].apply(
        lambda x: ', '.join(list(x)))
    fig = px.bar(sorted_df, x='antecedents',
                 y='confidence', color='consequents')
    st.plotly_chart(fig)