Differences

This shows you the differences between two versions of the page.

--- making_recommendation [2016/05/16 11:35] – created hkimscil
+++ making_recommendation [2017/03/13 12:55] (current) – hkimscil
@@ Line 1: / Line 1: @@
-{{:class:swap:recommendations_toby.pdf|Reading material}}
+{{c:swap:recommendations_toby.pdf|Reading material}}
+{{0y100685.pdf|상관관계를 이용한 recommendation system 예}}
+Python 실행
+<code py>c:\code\collective\chapter2> python
+Python 2.4.1 (#65, Mar 30 2005, 09:13:57) [MSC v.1310 32 bit (Intel)] on win32
+Type "help", "copyright", "credits" or "license" for more information.
+>>>
+</code>
+데이터 입력한다. 데이터는 critics라는 변수에 이름:영화:점수의 nested 형식으로 기록된다. recommendations.py에 저장
 <code py># A dictionary of movie critics and their ratings of a small
 # set of movies
@@ Line 27: / Line 34: @@
 'Toby': {'Snakes on a Plane':4.5,'You, Me and Dupree':1.0,'Superman Returns':4.0}}
 </code>
+Eucledian distance
+<imgcaption image1|people in preference space>{{people_in_preference_space.jpg |}}</imgcaption>
 <code>>> from math import sqrt
->> sqrt(pow(4.5-4,2)+pow(2-1,2))
+>> sqrt(pow(5-4,2)+pow(4-1,2))
-.118033988749895
+.1622776601683795
 </code>
 <code>>> 1/(1+sqrt(pow(5-4,2)+pow(4-1,2)))
-.4721359549995794
+.2402530733520421
 </code>
+<WRAP clear />
+아래는 persion1과 persion2 사이의 Distance를 구하는 definition
+위의 recommendations.py 에 추가.
 <code py>from math import sqrt
@@ Line 59: / Line 74: @@
 similarity between two users.
-<code>>>> reload(recommendations)
->>> recommendations.sim_distance(recommendations.critics,
+__for python 2.x__
-... 'Lisa Rose','Gene Seymour')
+<code>>>> import(recommendations)
+>>> # reload(recommendations)
+>>> recommendations.sim_distance(recommendations.critics,'Lisa Rose','Gene Seymour')
 .148148148148
 </code>
-for python 3.xx
+//for python 3.xx//
 <code>>>> import recommendations
 imp.reload(recommendations)
->>> recommendations.sim_distance(recommendations.critics,
+>>> recommendations.sim_distance(recommendations.critics,'Lisa Rose','Gene Seymour')
-... 'Lisa Rose','Gene Seymour')
 .148148148148
 </code>
@@ Line 112: / Line 128: @@
 <code>>>> reload(recommendations)
->>> print recommendations.sim_pearson(recommendations.critics,
+>>> print recommendations.sim_pearson(recommendations.critics,'Lisa Rose','Gene Seymour')
-... 'Lisa Rose','Gene Seymour')
+>>> # for python 3.4 아랫줄로 대체해서 사용할 것
+>>> print (recommendations.sim_pearson(recommendations.critics,'Lisa Rose','Gene Seymour'))
 .396059017191
 </code>
@@ Line 131: / Line 148: @@
 <code>>> reload(recommendations)
 >> recommendations.topMatches(recommendations.critics,'Toby',n=3)
-[(0.99124070716192991, 'Lisa Rose'), (0.92447345164190486, 'Mick LaSalle'),
+[(0.99124070716192991, 'Lisa Rose'), (0.92447345164190486, 'Mick
-(0.89340514744156474, 'Claudia Puig')]
+LaSalle'), (0.89340514744156474, 'Claudia Puig')]
 </code>
 ^ Critic  ^ Similarity  ^ Night  ^ S.xNight  ^ Lady  ^ S.xLady  ^ Luck  ^ S.xLuck  ^
@@ Line 182: / Line 197: @@
 <code>>>> reload(recommendations)
 >>> recommendations.getRecommendations(recommendations.critics,'Toby')
-[(3.3477895267131013, 'The Night Listener'), (2.8325499182641614, 'Lady in the
+[(3.3477895267131013, 'The Night Listener'), (2.8325499182641614, 'Lady in the Water'), (2.5309807037655645, 'Just My Luck')]
-Water'), (2.5309807037655645, 'Just My Luck')]
+>>> recommendations.getRecommendations(recommendations.critics,'Toby', similarity=recommendations.sim_distance)
->>> recommendations.getRecommendations(recommendations.critics,'Toby',
+[(3.5002478401415877, 'The Night Listener'), (2.7561242939959363, 'Lady in the Water'), (2.4619884860743739, 'Just My Luck')]
-... similarity=recommendations.sim_distance)
-[(3.5002478401415877, 'The Night Listener'), (2.7561242939959363, 'Lady in the
-Water'), (2.4619884860743739, 'Just My Luck')]
 </code>
@@ Line 348: / Line 360: @@
     c+=1
     if c%100==0: print "%d / %d" % (c,len(itemPrefs))
+    # for python3.4 윗줄을 아랫줄로 대체
+    # if c%100==0: print ("%d / %d" % (c,len(itemPrefs)))
     # Find the most similar items to this one
     scores=topMatches(itemPrefs,item,n=n,similarity=sim_distance)