recomm_refine.py

import pandas as pd

df = pd.read_excel('C:\\Users\\This PC\\Desktop\\rec.xlsx')

print(df)

print('For 10th level schooling -> 1','For 12th level schooling -> 2','For any technical diploma -> 3','For bachelor degree -> 4','For master degree -> 5',sep='\n')
print('For Ph.D -> 6')
user_edu = int(input('input your max edu code'))

print('For no degree -> 1','For degree in Technology -> 2','For degree in medical ->3','For degree in arts -> 4', 'For minimum any bachelor degree -> 5',sep = '\n')

user_stream = int(input('input your branch code'))


is_edu = df['education']==user_edu

df1 = df[is_edu]

print(df1)

df2 = df1.copy().loc[df['branch'] == user_stream]

print(df2)

m = df2['no_of_ratings'].quantile(0.90)

mf = df2['no_of_ratings'] > m

print(m)

print('recommended jobs')

print(df2[mf])

print('recommende jobs')

print('initializing-------------------------------------------------------')

C = df['ratings'].mean()

def weighted_rating(x, m=m, C=C):
    
    N = x['no_of_ratings']
    
    S = x['ratings']
    return (N/(N+m)*S) + (m/(m+N)*C)
    


df2['score'] = df2.apply(weighted_rating, axis=1)
#print(df2)

df2 = df2.sort_values('score', ascending=False)


print(df2[['job_name','score']])