Recommendation Algorithm

import pandas as pd
from sklearn.neighbors import NearestNeighbors


df = pd.read_csv('2020-4-27.csv')
df.columns
df.fillna(0, inplace=True)
df_norm = (df[features] - df[features].mean()) / df[features].std()
knn_model = NearestNeighbors(n_neighbors=5, algorithm='ball_tree')
knn_model.fit(df_norm)


def get_recommendations(query):
    features = ["Price", "Bedroom", "Bathroom", "Floors", "Parking", "Year"]
    query_df = pd.DataFrame(query, index=[0])
    query_norm = (query_df[features] - df[features].mean()) / df[features].std()
    distances, indices = knn_model.kneighbors(query_norm)
    recommended_properties = df.loc[indices[0]][["Title", "Address", "Price","Area"]]
    return recommended_properties


query = {"Price": 20000000, "Bedroom": 3, "Bathroom": 4, "Floors": 2, "Parking": 1, "Year": 2013}
recommended_properties = get_recommendations(query)
print(recommended_properties)