chapter4:內容過濾及分類－－－基於物品屬性的過濾

CopperDong發表於2017-10-05
APT

　　協同過濾也稱為社會過濾，利用了使用者社群的力量來幫助進行推薦，它的難點，包括資料稀疏和擴充套件性帶來的問題，另一個問題是基於協同過濾的推薦系統傾向於推薦已流行的物品，即偏向於流行事物。作為一個極端的例子，考慮一個全新樂隊剛釋出的專輯，由於樂隊和專輯從沒被人評價過或者沒人購買過，因此它永遠不會被推薦，這就是所謂的“冷啟動”問題。會帶來“富者越富”的效果

　　一種不同的推薦方法。考慮流音樂網站Pandora的推薦，基於一種稱為音樂基因的專案。他們僱了一些具有很強音樂理論背景的專業音樂人士作為分析師，有他們來決定歌曲的特徵（他們稱之為基因）。這些分析師會接受超過１５０個小時的培訓。一旦培訓完畢，他們就會花平均２０～３０分鐘的時間來分析一首歌曲以確定其基因或者說特徵。這些特徵當中很多都是專業性的。分析師會在超過４００中基因上進行評分。由於每個月都大約新增１５０００首新歌，因此上述做法的工作量很大。

一、選擇合適取值的重要性

　　特徵選取，如音樂的流派、情緒，取值在１～５之間

用Python實現的資料格式

music = {"Dr Dog/Fate": {"piano": 2.5, "vocals": 4, "beat": 3.5, "blues": 3, "guitar": 5, "backup vocals": 4, "rap": 1},
         "Phoenix/Lisztomania": {"piano": 2, "vocals": 5, "beat": 5, "blues": 3, "guitar": 2, "backup vocals": 1, "rap": 1},
         "Heartless Bastards/Out at Sea": {"piano": 1, "vocals": 5, "beat": 4, "blues": 2, "guitar": 4, "backup vocals": 1, "rap": 1},
         "Todd Snider/Don't Tempt Me": {"piano": 4, "vocals": 5, "beat": 4, "blues": 4, "guitar": 1, "backup vocals": 5, "rap": 1},
         "The Black Keys/Magic Potion": {"piano": 1, "vocals": 4, "beat": 5, "blues": 3.5, "guitar": 5, "backup vocals": 1, "rap": 1},
         "Glee Cast/Jessie's Girl": {"piano": 1, "vocals": 5, "beat": 3.5, "blues": 3, "guitar":4, "backup vocals": 5, "rap": 1},
         "La Roux/Bulletproof": {"piano": 5, "vocals": 5, "beat": 4, "blues": 2, "guitar": 1, "backup vocals": 1, "rap": 1},
         "Mike Posner": {"piano": 2.5, "vocals": 4, "beat": 4, "blues": 1, "guitar": 1, "backup vocals": 1, "rap": 1},
         "Black Eyed Peas/Rock That Body": {"piano": 2, "vocals": 5, "beat": 5, "blues": 1, "guitar": 2, "backup vocals": 2, "rap": 4},
         "Lady Gaga/Alejandro": {"piano": 1, "vocals": 5, "beat": 3, "blues": 2, "guitar": 1, "backup vocals": 2, "rap": 1}}
用曼哈頓距離推薦 

from math import sqrt

users = {"Angelica": {"Blues Traveler": 3.5, "Broken Bells": 2.0, "Norah Jones": 4.5, "Phoenix": 5.0, "Slightly Stoopid": 1.5, "The Strokes": 2.5, "Vampire Weekend": 2.0},
         "Bill":{"Blues Traveler": 2.0, "Broken Bells": 3.5, "Deadmau5": 4.0, "Phoenix": 2.0, "Slightly Stoopid": 3.5, "Vampire Weekend": 3.0},
         "Chan": {"Blues Traveler": 5.0, "Broken Bells": 1.0, "Deadmau5": 1.0, "Norah Jones": 3.0, "Phoenix": 5, "Slightly Stoopid": 1.0},
         "Dan": {"Blues Traveler": 3.0, "Broken Bells": 4.0, "Deadmau5": 4.5, "Phoenix": 3.0, "Slightly Stoopid": 4.5, "The Strokes": 4.0, "Vampire Weekend": 2.0},
         "Hailey": {"Broken Bells": 4.0, "Deadmau5": 1.0, "Norah Jones": 4.0, "The Strokes": 4.0, "Vampire Weekend": 1.0},
         "Jordyn":  {"Broken Bells": 4.5, "Deadmau5": 4.0, "Norah Jones": 5.0, "Phoenix": 5.0, "Slightly Stoopid": 4.5, "The Strokes": 4.0, "Vampire Weekend": 4.0},
         "Sam": {"Blues Traveler": 5.0, "Broken Bells": 2.0, "Norah Jones": 3.0, "Phoenix": 5.0, "Slightly Stoopid": 4.0, "The Strokes": 5.0},
         "Veronica": {"Blues Traveler": 3.0, "Norah Jones": 5.0, "Phoenix": 4.0, "Slightly Stoopid": 2.5, "The Strokes": 3.0}
        }

music = {"Dr Dog/Fate": {"piano": 2.5, "vocals": 4, "beat": 3.5, "blues": 3, "guitar": 5, "backup vocals": 4, "rap": 1},
         "Phoenix/Lisztomania": {"piano": 2, "vocals": 5, "beat": 5, "blues": 3, "guitar": 2, "backup vocals": 1, "rap": 1},
         "Heartless Bastards/Out at Sea": {"piano": 1, "vocals": 5, "beat": 4, "blues": 2, "guitar": 4, "backup vocals": 1, "rap": 1},
         "Todd Snider/Don't Tempt Me": {"piano": 4, "vocals": 5, "beat": 4, "blues": 4, "guitar": 1, "backup vocals": 5, "rap": 1},
         "The Black Keys/Magic Potion": {"piano": 1, "vocals": 4, "beat": 5, "blues": 3.5, "guitar": 5, "backup vocals": 1, "rap": 1},
         "Glee Cast/Jessie's Girl": {"piano": 1, "vocals": 5, "beat": 3.5, "blues": 3, "guitar":4, "backup vocals": 5, "rap": 1},
         "La Roux/Bulletproof": {"piano": 5, "vocals": 5, "beat": 4, "blues": 2, "guitar": 1, "backup vocals": 1, "rap": 1},
         "Mike Posner": {"piano": 2.5, "vocals": 4, "beat": 4, "blues": 1, "guitar": 1, "backup vocals": 1, "rap": 1},
         "Black Eyed Peas/Rock That Body": {"piano": 2, "vocals": 5, "beat": 5, "blues": 1, "guitar": 2, "backup vocals": 2, "rap": 4},
         "Lady Gaga/Alejandro": {"piano": 1, "vocals": 5, "beat": 3, "blues": 2, "guitar": 1, "backup vocals": 2, "rap": 1}}

def manhattan(rating1, rating2):
    """Computes the Manhattan distance. Both rating1 and rating2 are dictionaries
       of the form {'The Strokes': 3.0, 'Slightly Stoopid': 2.5}"""
    distance = 0
    total = 0
    for key in rating1:
        if key in rating2:
            distance += abs(rating1[key] - rating2[key])
            total += 1
    return distance



def computeNearestNeighbor(username, users):
    """creates a sorted list of users based on their distance to username"""
    distances = []
    for user in users:
        if user != username:
            distance = manhattan(users[user], users[username])
            distances.append((distance, user))
    # sort based on distance -- closest first
    distances.sort()
    return distances

def recommend(username, users):
    """Give list of recommendations"""
    # first find nearest neighbor
    nearest = computeNearestNeighbor(username, users)[0][1]

    recommendations = []
    # now find bands neighbor rated that user didn't
    neighborRatings = users[nearest]
    userRatings = users[username]
    for artist in neighborRatings:
        if not artist in userRatings:
            recommendations.append((artist, neighborRatings[artist]))
    # using the fn sorted for variety - sort is more efficient
    return sorted(recommendations, key=lambda artistTuple: artistTuple[1], reverse = True)
一個取值範圍的問題

　　假設某個特徵在距離計算中占主導地位，並不是什麼好事，實際上，這種不同屬性取值範圍的差異對任意推薦系統來說都是個大問題。

二、歸一化

　　解決上面的問題是歸一化。為了消除資料的偏斜性，我們必須要對資料標準化或者說歸一化。

　　一個常用的歸一化方法會將每個特徵的值轉換為０到１之間，如　(val - min) / (max - min) 

       如果你上過統計課，可能會熟悉更精確的標準化資料的做法，如標準分數(Standard Score)

      

　　使用標準分數的問題在於其會受到離群點的劇烈影響。

改進的標準分數



哪些情況下應該進行歸一化處理：記住的是如果進行歸一化的話會涉及計算的開銷

　　１、所用資料探勘方法基於特徵的值來計算兩個物件的距離

　　２、不同特徵的尺度不同（特別是有顯著不同的情況，如上述例子中的詢價和臥室數目）


三、最近鄰分類器的Python程式碼

      為喜歡Green　Day的使用者推薦歌曲

　　需要的資料

　　　　音樂的屬性music = { }

              將music轉換成向量items = { }　方便計算

              每個使用者對部分的評分users = { }

　　建立一個分類函式


四、體育專案的識別

　　小規模資料，兩個檔案athletesTrainingSet.txt（訓練分類器） and athletesTestSet.txt（評估分類器）

class Classifier:

    def __init__(self, filename):

        self.medianAndDeviation = []
        
        # reading the data in from the file
        f = open(filename)
        lines = f.readlines()
        f.close()
        self.format = lines[0].strip().split('\t')
        self.data = []
        for line in lines[1:]:
            fields = line.strip().split('\t')
            ignore = []
            vector = []
            for i in range(len(fields)):
                if self.format[i] == 'num':
                    vector.append(int(fields[i]))
                elif self.format[i] == 'comment':
                    ignore.append(fields[i])
                elif self.format[i] == 'class':
                    classification = fields[i]
            self.data.append((classification, vector, ignore))
        self.rawData = list(self.data)
        

        
    
    ##################################################
    ###
    ###  FINISH THE FOLLOWING TWO METHODS

    def getMedian(self, alist):
        """return median of alist"""

        """TO BE DONE"""
        return 0
        

    def getAbsoluteStandardDeviation(self, alist, median):
        """given alist and median return absolute standard deviation"""

        """TO BE DONE"""
        return 0

    
    ###
    ### 
    ##################################################



def unitTest():
    list1 = [54, 72, 78, 49, 65, 63, 75, 67, 54]
    list2 = [54, 72, 78, 49, 65, 63, 75, 67, 54, 68]
    list3 = [69]
    list4 = [69, 72]
    classifier = Classifier('athletesTrainingSet.txt')
    m1 = classifier.getMedian(list1)
    m2 = classifier.getMedian(list2)
    m3 = classifier.getMedian(list3)
    m4 = classifier.getMedian(list4)
    asd1 = classifier.getAbsoluteStandardDeviation(list1, m1)
    asd2 = classifier.getAbsoluteStandardDeviation(list2, m2)
    asd3 = classifier.getAbsoluteStandardDeviation(list3, m3)
    asd4 = classifier.getAbsoluteStandardDeviation(list4, m4)
    assert(round(m1, 3) == 65)
    assert(round(m2, 3) == 66)
    assert(round(m3, 3) == 69)
    assert(round(m4, 3) == 70.5)
    assert(round(asd1, 3) == 8)
    assert(round(asd2, 3) == 7.5)
    assert(round(asd3, 3) == 0)
    assert(round(asd4, 3) == 1.5)
    
    print("getMedian and getAbsoluteStandardDeviation work correctly")

unitTest()

五、Iris資料集


六、汽車ＭＰＧ資料

　　該資料來自卡內基梅隆大學，最初用於１９８３年度的美國統計協會展會上。

七、雜談

　　　注意歸一化，重要性





相關文章