Initial

collab-filter.py

+import csv
+
+csvfile=open('movie-ratings.csv')
+read=csv.reader(csvfile)
+reading=[]
+for line in read:
+	reading.append(line)
+f2 = open ("user_preference.csv")
+data = csv.reader(f2)
+preference = []
+a=input("Your name :")
+preference.append(a)
+for line in data:
+    preference.append(line)
+prefdict = dict( zip(preference[1],preference[2]))
+print(prefdict)
\ No newline at end of file

movie-ratings.csv

+Name,The Godfather,Pulp Fiction,Gone With the Wind,12 Angry Men,Lawrence of Arabia,Raging Bull,The Clockwork Orange,Cinema Paradiso,Wild Strawberries,Amadeus
+Albert Einstein,9.5,7,10,10,-1,-1,6,10,8,10
+The Joker,8,10,4,4,8,-1,10,9.5,7.5,8
+Marilyn Monroe,8,-1,10,8.5,-1,8.5,6.5,-1,8,5
+Steve Jobs,-1,8.5,8,10,9,7.5,9,8,8.5,9
+Mahatma Gandhi,6,-1,-1,9,8,8,5,7,7,7
+Pikachu,7,-1,5,-1,-1,9,-1,5,-1,10
+Jon Snow,-1,7,10,8,8,8,8,8,8,-1
+Walter White,9,10,6,6,7,-1,10,7,10,-1
+J K Rowling,9,9,-1,9,8,7,9,10,10,9
+Diego Maradona,-1,8,7,6,7,9,6,8,7,-1
+Barrack Obama,7,9,10,10,7,9,-1,-1,-1,8

temp.py

+
+
+import csv
+
+f1 = open ("movie-ratings.csv" , "rb")
+
+data = csv.reader(f1)
+
+rating = []
+
+for line in data:
+
+    rating.append(line)
+
+    f2 = open ("user_preference.csv" , "rb")
+
+    data = csv.reader(f2)
+
+    preference = []
+
+    preference.append(input("Your name :"))
+
+for line in data:
+
+    preference.append(line)
+
+prefdict = dict( zip(preference[1],preference[2]))
+
+
+header = rating[0]
+
+del rating[0]
+
+reviewers = [ s[0] for s in rating]
+
+reviewdic = []
+
+for line in rating:
+
+    dic = dict( (header[i],line[i]) for i in range(1,len(line)))
+
+    reviewdic.append(dic)
+
+
+
+from math import sqrt
+
+euc = []
+
+for i in reviewdic:
+
+    diff = [(float(i[s])-float(prefdict[s]))**2 for s in prefdict.keys()]
+
+    euc.append(sqrt(sum(diff)))
+
+ordereuc = sorted(zip(reviewers,euc), key=lambda x: x[1])
+
+
+
+pear = []
+
+x = [float(i) for i in prefdict.values()]
+
+x2 = [ i**2 for i in x]
+
+n = len(x)
+
+for i in reviewdic:
+
+    y = [float(i[j]) for j in prefdict.keys()]
+
+    y2 = [ i**2 for i in y]
+
+    xy = [ x[i]*y[i] for i in range(n)]
+
+    corr = (n*sum(xy) - sum(x)*sum(y))/(sqrt(n*sum(x2) - sum(x)**2)*sqrt(n*sum(y2) - sum(y)**2))
+
+    pear.append(corr)
+
+orderpear = sorted(zip(reviewers,pear), key=lambda x: x[1], reverse = True)
+
+
+
+
+finalrating = {}
+
+for movie in header[1:]:
+
+    finalrating[movie] = sum([pear[i]*float(reviewdic[i][movie]) for i in range(len(pear))])
+
+
+
+finalorder = sorted(finalrating.items(), key=lambda x: x[1], reverse = True)
+
+
+towatch = []
+
+for movie,rating in finalorder:
+
+    if movie in prefdict.keys():
+
+        continue
+
+    else:
+
+        towatch.append(movie)
+
+
+        if len(towatch)>=3:
+
+            for i in range(3):
+
+                print (towatch[i])
+
+            else:
+
+                for movie,rating in finalorder:
+
+                    print (movie)
+

user_preference.csv

+The Godfather,Pulp Fiction,The Clockwork Orange,Cinema Paradiso,Wild Strawberries,Amadeus
+9,7,6,10,8,10