Add fuzzylogic aggression metric

data_filters.py

@@ -1,6 +1,7 @@
 import pandas as pd
 from simpful import *
 
+FS = FuzzySystem()
 def generateTrainingData(dataframe):
     columns = ['season','date','home_team','away_team','result_full','home_passes','away_passes',
                'home_possession','away_possession','home_shots','away_shots']
@@ -10,10 +11,61 @@ def generateTrainingData(dataframe):
 def generateFuzzyLogicData(dataframe):
     columns = ['season','date','home_team','away_team','result_full','c_home_form_5m','c_away_form_5m',#,'c_home_passes','c_away_passes',
               # 'c_home_possession','c_away_possession','c_home_shots','c_away_shots',
-               'c_home_diff_5m', 'c_away_diff_5m','c_home_form_season','c_away_form_season',
+               'c_home_diff_5m', 'c_away_diff_5m', 'c_home_aggression_5m',
+               'c_away_aggression_5m', 'c_home_aggression_season', 'c_away_aggression_season',
+               'c_home_form_season','c_away_form_season',
                'c_home_diff_season', 'c_away_diff_season']
     return dataframe[columns]
 
+def calculateFuzzyAggression(yellow_cards, red_cards):
+    FS.set_crisp_output_value("low", 0.0)
+    FS.set_crisp_output_value("average", 0.5)
+    FS.set_crisp_output_value("high", 1.0)
+
+    Yellow_cards1 = TriangleFuzzySet(0, 2, 3, term="low")
+    Yellow_cards2 = TriangleFuzzySet(2, 3, 4, term="average")
+    Yellow_cards3 = TriangleFuzzySet(3, 4, 4, term="high")
+
+    FS.add_linguistic_variable("yellow_cards", LinguisticVariable([Yellow_cards1, Yellow_cards2, Yellow_cards3], universe_of_discourse=[0, 10]))
+
+    Red_cards1 = TriangleFuzzySet(0, 0, 1, term="low")
+    Red_cards2 = TriangleFuzzySet(0, 1, 2, term="average")
+    Red_cards3 = TriangleFuzzySet(1, 2, 2, term="high")
+
+    FS.add_linguistic_variable("red_cards", LinguisticVariable([Red_cards1, Red_cards2, Red_cards3], universe_of_discourse=[0, 4]))
+
+    # Pass_domination1 = TriangleFuzzySet(2,2,6,   term="low")
+    # Pass_domination2 = TriangleFuzzySet(3,5,7,  term="average")
+    # Pass_domination3 = TriangleFuzzySet(4,8,8, term="high")
+
+    # FS.add_linguistic_variable("passes_domination", LinguisticVariable([Pass_domination1, Pass_domination2, Pass_domination3], universe_of_discourse=[0,10]))
+
+    FS.add_rules([
+
+        "IF (yellow_cards IS low) AND (red_cards IS low) THEN (aggression IS low)",
+
+        "IF (yellow_cards IS high) AND (red_cards IS high) THEN (aggression IS high)",
+
+        "IF (yellow_cards IS average) AND (red_cards IS average) THEN (aggression IS average)",
+
+        "IF (yellow_cards IS low) AND (red_cards IS high) THEN (aggression IS high)",
+
+        "IF (yellow_cards IS high ) AND (red_cards IS low) THEN (aggression IS high)",
+
+        "IF (yellow_cards IS average) AND (red_cards IS high) THEN (aggression IS high)",
+
+        "IF (yellow_cards IS high) AND (red_cards IS average) THEN (aggression IS high)",
+
+        "IF (yellow_cards IS low) AND (red_cards IS average) THEN (aggression IS low)",
+
+        "IF (yellow_cards IS average) AND (red_cards IS low) THEN (aggression IS average)"
+    ])
+
+    FS.set_variable("yellow_cards", yellow_cards)
+    FS.set_variable("red_cards", red_cards)
+
+    aggression = FS.inference()
+    return aggression
 def last5Matches(season, teamA, data, df):
     # Wybierz rekordy dla danej pary drużyn i sezonu
     subset = df[((df['season'] == season) & ((df['home_team'] == teamA) | (df['away_team'] == teamA)))]
@@ -54,6 +106,23 @@ def getResult(score,teamHome):
     else:
         return "loss"
 
+def calculateAggression(matches, team):
+    aggression = 0
+    for index, row in matches.iterrows():
+        if team == row['home_team']:
+            yellow_cards = row['home_yellow_cards']
+            red_cards = row['home_red_cards']
+        else:
+            yellow_cards = row['away_yellow_cards']
+            red_cards = row['away_red_cards']
+        aggression_result = calculateFuzzyAggression(yellow_cards, red_cards)
+        #print(aggression_result['aggression'])
+        aggression = aggression + aggression_result['aggression']
+    if matches.shape[0] != 0:
+        aggression_avg = aggression / matches.shape[0]
+    else:
+        aggression_avg = 0
+    return aggression_avg
 
 def calculatePoints(matches, team):
     points = 0

main.py

@@ -23,7 +23,6 @@ from sklearn.metrics import classification_report
 if __name__ == "__main__":
 
     df = pd.read_csv('df_full_premierleague.csv')
-
     result = last5Matches('10/11', 'Stoke City', '2010-10-02', df)
     #print(result.to_markdown())
     #print(result)
@@ -50,13 +49,17 @@ if __name__ == "__main__":
 
     df = add_column(df, get_method(df, True, categorize_diff, last5Matches), "c_home_diff_5m")#categorize_diff
     df = add_column(df, get_method(df, False, categorize_diff,last5Matches), "c_away_diff_5m")
-
+    df = add_column(df, get_method(df, True, categorize_aggression, last5Matches), "c_home_aggression_5m")#categorize_diff
+    df = add_column(df, get_method(df, False, categorize_aggression,last5Matches), "c_away_aggression_5m")
     df = add_column(df, get_method(df, True, categorize_points, seasonMatches), "c_home_form_season")
     df = add_column(df, get_method(df, False, categorize_points, seasonMatches), "c_away_form_season")
 
     df = add_column(df, get_method(df, True, categorize_diff, seasonMatches), "c_home_diff_season")#categorize_diff
     df = add_column(df, get_method(df, False, categorize_diff,seasonMatches), "c_away_diff_season")
 
+    df = add_column(df, get_method(df, True, categorize_aggression, seasonMatches), "c_home_aggression_season")#categorize_diff
+    df = add_column(df, get_method(df, False, categorize_aggression,seasonMatches), "c_away_aggression_season")
+
     df = generateFuzzyLogicData(df)
 
     label_encoder = LabelEncoder()

rules.py

@@ -190,7 +190,18 @@ def categorize_diff(data, row, teamHome, matches_type):
     else:
         return 1
 
-
+def categorize_aggression(data, row, teamHome, matches_type):
+    if teamHome:
+        data_5 = matches_type(row['season'], row['home_team'], row['date'], data)
+        diff = calculateAggression(data_5,row['home_team'])
+    else:
+        data_5 = matches_type(row['season'], row['away_team'], row['date'], data)
+        diff = calculateAggression(data_5,row['away_team'])
+    return diff
+   # if diff <=0:
+    #    return 0
+   # else:
+   #     return 1
 def get_diff_home(data):
     points = []
     for index, row in data.iterrows():