Merge pull request #181 from sanskriti-lal/add-sanskriti-lal

added an  AI-powered system to match food donors with recipients

matching-model/convertDStoCSV.py

@@ -0,0 +1,22 @@
+# Converts the datatset into csv file - i.e the donations table is being converted into .csv file
+
+import pandas as pd
+from your_flask_app import app, db # Import your Flask app and SQLAlchemy objects
+from models import Donation # Import the Donation model
+
+def export_to_csv():
+ with app.app_context():
+ donations = Donation.query.all()
+ data = [{
+ 'food_type': donation.food_type,
+ 'quantity': donation.quantity,
+ 'location': donation.location,
+ 'timestamp': donation.timestamp
+ } for donation in donations]
+
+ df = pd.DataFrame(data)
+ df.to_csv('donations_requests.csv', index=False)
+
+# Call this function periodically, e.g., using a cron job or a scheduled task
+export_to_csv()
+

matching-model/main.py

@@ -0,0 +1,50 @@
+import pandas as pd
+from sklearn.model_selection import train_test_split
+from sklearn.neighbors import KNeighborsClassifier
+from sklearn.metrics.pairwise import cosine_similarity
+
+# Load the CSV file into a pandas DataFrame (assuming data is already loaded)
+df = pd.read_csv('convertDStoCSV.csv')
+
+# Preprocess the data (e.g., handle missing values, encode categorical variables)
+# Assuming 'timestamp' is in string format, convert it to datetime if needed
+df['timestamp'] = pd.to_datetime(df['timestamp'])
+
+# Split the data into features (X) and target variable (y)
+X = df[['food_type', 'location', 'quantity', 'timestamp']] # Features: food_type, location, quantity, timestamp
+y = df['food_preferences'] # Target variable: food_preferences
+
+# Encode categorical variables (if needed)
+X_encoded = pd.get_dummies(X, columns=['food_type', 'location'], drop_first=True) # One-hot encoding for categorical variables
+
+# Split the data into training and testing sets
+X_train, X_test, y_train, y_test = train_test_split(X_encoded, y, test_size=0.2, random_state=42)
+
+# Initialize the KNN classifier
+knn = KNeighborsClassifier(n_neighbors=3) # Adjust hyperparameters as needed
+
+# Train the classifier
+knn.fit(X_train, y_train)
+
+# Calculate similarity between donors and recipients
+def calculate_similarity(row):
+ donor_food = row['food_type'].split(', ')
+ recipient_food = row['food_preferences'].split(', ')
+ vector_donor = [1 if food in donor_food else 0 for food in set(donor_food + recipient_food)]
+ vector_recipient = [1 if food in recipient_food else 0 for food in set(donor_food + recipient_food)]
+ similarity = cosine_similarity([vector_donor], [vector_recipient])[0][0]
+ return similarity
+
+# Calculate similarity scores for the test set
+X_test_similarity = X_test.apply(calculate_similarity, axis=1)
+
+# Set a similarity threshold to filter matches
+threshold = 0.5
+matched_indices = X_test_similarity[X_test_similarity >= threshold].index
+
+# Get the matched donor-recipient pairs
+matched_pairs = df.iloc[matched_indices]
+
+# Output matched pairs along with their similarity scores
+print("Matched Donor-Recipient Pairs:")
+print(matched_pairs[['donor_id', 'recipient_id', 'location', 'quantity']])