当前位置：首页 > news >正文

数据分析作业四-基于用户及物品数据进行内容推荐

news 2025/6/22 14:27:34

## 导入支持库
import pandas as pd
import matplotlib.pyplot as plt
import sklearn.metrics as metrics
import numpy as np
from sklearn.neighbors import NearestNeighbors
from scipy.spatial.distance import correlation
from sklearn.metrics.pairwise import pairwise_distances
import ipywidgets as widgets
from IPython.display import display, clear_output
from contextlib import contextmanager
import warnings
warnings.filterwarnings('ignore')
import numpy as np
import os, sys
import re
import seaborn as sns

## 加载数据集并检查书籍，用户和评级数据集的形状
books = pd.read_csv('F:\\data\\bleeding_data\\BX-Books.csv',sep=None,encoding="latin-1")
books.columns = ['ISBN', 'bookTitle', 'bookAuthor','yearOfPublication', 'publisher','imageUrlS', 'imageUrlM', 'imageUrlL']users = pd.read_csv('F:\\data\\bleeding_data\\BX-Users.csv',sep=None, encoding="latin-1")
users.columns = ['userID', 'Location', 'Age']ratings = pd.read_csv('F:\\data\\bleeding_data\\BX-Book-Ratings.csv',sep=None, encoding="latin-1")
ratings.columns = ['userID', 'ISBN', 'bookRating']print (books.shape)
print (users.shape)
print (ratings.shape)

(271360, 8)
(278858, 3)
(1149780, 3)

## 一、图书数据集
books.head()

	ISBN	bookTitle	bookAuthor	yearOfPublication	publisher	imageUrlS	imageUrlM	imageUrlL
0	0195153448	Classical Mythology	Mark P. O. Morford	2002	Oxford University Press	http://images.amazon.com/images/P/0195153448.0...	http://images.amazon.com/images/P/0195153448.0...	http://images.amazon.com/images/P/0195153448.0...
1	0002005018	Clara Callan	Richard Bruce Wright	2001	HarperFlamingo Canada	http://images.amazon.com/images/P/0002005018.0...	http://images.amazon.com/images/P/0002005018.0...	http://images.amazon.com/images/P/0002005018.0...
2	0060973129	Decision in Normandy	Carlo D'Este	1991	HarperPerennial	http://images.amazon.com/images/P/0060973129.0...	http://images.amazon.com/images/P/0060973129.0...	http://images.amazon.com/images/P/0060973129.0...
3	0374157065	Flu: The Story of the Great Influenza Pandemic...	Gina Bari Kolata	1999	Farrar Straus Giroux	http://images.amazon.com/images/P/0374157065.0...	http://images.amazon.com/images/P/0374157065.0...	http://images.amazon.com/images/P/0374157065.0...
4	0393045218	The Mummies of Urumchi	E. J. W. Barber	1999	W. W. Norton & Company	http://images.amazon.com/images/P/0393045218.0...	http://images.amazon.com/images/P/0393045218.0...	http://images.amazon.com/images/P/0393045218.0...

## url不需要分析，进行删除
books.drop(['imageUrlS', 'imageUrlM', 'imageUrlL'],axis=1,inplace=True)
books.head()

	ISBN	bookTitle	bookAuthor	yearOfPublication	publisher
0	0195153448	Classical Mythology	Mark P. O. Morford	2002	Oxford University Press
1	0002005018	Clara Callan	Richard Bruce Wright	2001	HarperFlamingo Canada
2	0060973129	Decision in Normandy	Carlo D'Este	1991	HarperPerennial
3	0374157065	Flu: The Story of the Great Influenza Pandemic...	Gina Bari Kolata	1999	Farrar Straus Giroux
4	0393045218	The Mummies of Urumchi	E. J. W. Barber	1999	W. W. Norton & Company

## books.dtypes
books.dtypes

ISBN                 object
bookTitle            object
bookAuthor           object
yearOfPublication    object
publisher            object
dtype: object

## 现在检查属性的唯一值
books.bookTitle.unique()

array(['Classical Mythology', 'Clara Callan', 'Decision in Normandy', ...,'Lily Dale : The True Story of the Town that Talks to the Dead',"Republic (World's Classics)","A Guided Tour of Rene Descartes' Meditations on First Philosophy with Complete Translations of the Meditations by Ronald Rubin"],dtype=object)

books.yearOfPublication.unique()

array(['2002', '2001', '1991', '1999', '2000', '1993', '1996', '1988','2004', '1998', '1994', '2003', '1997', '1983', '1979', '1995','1982', '1985', '1992', '1986', '1978', '1980', '1952', '1987','1990', '1981', '1989', '1984', '0', '1968', '1961', '1958','1974', '1976', '1971', '1977', '1975', '1965', '1941', '1970','1962', '1973', '1972', '1960', '1966', '1920', '1956', '1959','1953', '1951', '1942', '1963', '1964', '1969', '1954', '1950','1967', '2005', '1957', '1940', '1937', '1955', '1946', '1936','1930', '2011', '1925', '1948', '1943', '1947', '1945', '1923','2020', '1939', '1926', '1938', '2030', '1911', '1904', '1949','1932', '1928', '1929', '1927', '1931', '1914', '2050', '1934','1910', '1933', '1902', '1924', '1921', '1900', '2038', '2026','1944', '1917', '1901', '2010', '1908', '1906', '1935', '1806','2021', '2012', '2006', 'DK Publishing Inc', 'Gallimard', '1909','2008', '1378', '1919', '1922', '1897', '2024', '1376', '2037'],dtype=object)

books.loc[books.yearOfPublication == 'DK Publishing Inc',:]
books.yearOfPublication.unique()

array(['2002', '2001', '1991', '1999', '2000', '1993', '1996', '1988','2004', '1998', '1994', '2003', '1997', '1983', '1979', '1995','1982', '1985', '1992', '1986', '1978', '1980', '1952', '1987','1990', '1981', '1989', '1984', '0', '1968', '1961', '1958','1974', '1976', '1971', '1977', '1975', '1965', '1941', '1970','1962', '1973', '1972', '1960', '1966', '1920', '1956', '1959','1953', '1951', '1942', '1963', '1964', '1969', '1954', '1950','1967', '2005', '1957', '1940', '1937', '1955', '1946', '1936','1930', '2011', '1925', '1948', '1943', '1947', '1945', '1923','2020', '1939', '1926', '1938', '2030', '1911', '1904', '1949','1932', '1928', '1929', '1927', '1931', '1914', '2050', '1934','1910', '1933', '1902', '1924', '1921', '1900', '2038', '2026','1944', '1917', '1901', '2010', '1908', '1906', '1935', '1806','2021', '2012', '2006', 'DK Publishing Inc', 'Gallimard', '1909','2008', '1378', '1919', '1922', '1897', '2024', '1376', '2037'],dtype=object)

print(books.loc[books.yearOfPublication == 'DK Publishing Inc',:])

              ISBN                                          bookTitle  \
209538  078946697X  DK Readers: Creating the X-Men, How It All Beg...   
221678  0789466953  DK Readers: Creating the X-Men, How Comic Book...   bookAuthor  yearOfPublication  \
209538       2000  DK Publishing Inc   
221678       2000  DK Publishing Inc   publisher  
209538  http://images.amazon.com/images/P/078946697X.0...  
221678  http://images.amazon.com/images/P/0789466953.0...

books.loc[books.yearOfPublication == 'DK Publishing Inc',:]

	ISBN	bookTitle	bookAuthor	yearOfPublication	publisher
209538	078946697X	DK Readers: Creating the X-Men, How It All Beg...	2000	DK Publishing Inc	http://images.amazon.com/images/P/078946697X.0...
221678	0789466953	DK Readers: Creating the X-Men, How Comic Book...	2000	DK Publishing Inc	http://images.amazon.com/images/P/0789466953.0...

## 从上面可以看出，bookAuthor错误地装载了bookTitle，因此需要进行修正。
# ISBN '0789466953'
books.loc[books.ISBN == '0789466953','yearOfPublication'] = 2000
books.loc[books.ISBN == '0789466953','bookAuthor'] = "James Buckley"
books.loc[books.ISBN == '0789466953','publisher'] = "DK Publishing Inc"
books.loc[books.ISBN == '0789466953','bookTitle'] = "DK Readers: Creating the X-Men, How Comic Books Come to Life (Level 4: Proficient Readers)"#ISBN '078946697X'
books.loc[books.ISBN == '078946697X','yearOfPublication'] = 2000
books.loc[books.ISBN == '078946697X','bookAuthor'] = "Michael Teitelbaum"
books.loc[books.ISBN == '078946697X','publisher'] = "DK Publishing Inc"
books.loc[books.ISBN == '078946697X','bookTitle'] = "DK Readers: Creating the X-Men, How It All Began (Level 4: Proficient Readers)"

books.loc[(books.ISBN == '0789466953') | (books.ISBN == '078946697X'),:]

	ISBN	bookTitle	bookAuthor	yearOfPublication	publisher
209538	078946697X	DK Readers: Creating the X-Men, How It All Beg...	Michael Teitelbaum	2000	DK Publishing Inc
221678	0789466953	DK Readers: Creating the X-Men, How Comic Book...	James Buckley	2000	DK Publishing Inc

## 继续纠正出版年鉴的类型
books.yearOfPublication=pd.to_numeric(books.yearOfPublication, errors='coerce')
sorted(books['yearOfPublication'].unique())

[0.0,1376.0,1378.0,1806.0,1897.0,1900.0,1901.0,1902.0,1904.0,1906.0,1908.0,1909.0,1910.0,1911.0,1914.0,1917.0,1919.0,1920.0,1921.0,1922.0,1923.0,1924.0,1925.0,1926.0,1927.0,1928.0,1929.0,1930.0,1931.0,1932.0,1933.0,1934.0,1935.0,1936.0,1937.0,1938.0,1939.0,1940.0,1941.0,1942.0,1943.0,1944.0,1945.0,1946.0,1947.0,1948.0,1949.0,1950.0,1951.0,1952.0,1953.0,1954.0,1955.0,1956.0,1957.0,1958.0,1959.0,1960.0,1961.0,1962.0,1963.0,1964.0,1965.0,1966.0,1967.0,1968.0,1969.0,1970.0,1971.0,1972.0,1973.0,1974.0,1975.0,1976.0,1977.0,1978.0,1979.0,1980.0,1981.0,1982.0,1983.0,1984.0,1985.0,1986.0,1987.0,1988.0,1989.0,1990.0,1991.0,1992.0,1993.0,1994.0,1995.0,1996.0,1997.0,1998.0,1999.0,2000.0,2001.0,2002.0,2003.0,2004.0,2005.0,2006.0,2008.0,2010.0,2011.0,2012.0,2020.0,2021.0,2024.0,2026.0,2030.0,2037.0,2038.0,2050.0,nan]

## 现在可以看出yearOfPublication的类型为int，其值范围为0-2050。## 由于该数据集建于2004年，我假设2006年之后的所有年份都无效，保留两年的保证金，以防数据集可能已更新。## 对于所有无效条目（包括0），我将这些条目转换为NaN，然后用剩余年份的平均值替换它们。
books.loc[(books.yearOfPublication > 2006) | (books.yearOfPublication == 0),'yearOfPublication'] = np.NAN

# 用年出版的平均价值代替NaNs在案例数据集被更新的情况下保留一定的空白
books.yearOfPublication.fillna(round(books.yearOfPublication.mean()), inplace=True)
books.yearOfPublication.isnull().sum()

books.yearOfPublication = books.yearOfPublication.astype(np.int32)

## publisher
books.loc[books.publisher.isnull(),:]

	ISBN	bookTitle	bookAuthor	yearOfPublication	publisher
128890	193169656X	Tyrant Moon	Elaine Corvidae	2002	NaN
129037	1931696993	Finders Keepers	Linnea Sinclair	2001	NaN

## 检查行是否有书签作为查找器，看看我们是否能得到任何线索## 与不同的出版商和图书作者的所有行
books.loc[(books.bookTitle == 'Tyrant Moon'),:]

	ISBN	bookTitle	bookAuthor	yearOfPublication	publisher
128890	193169656X	Tyrant Moon	Elaine Corvidae	2002	NaN

books.loc[(books.bookTitle == 'Finders Keepers'),:]

	ISBN	bookTitle	bookAuthor	yearOfPublication	publisher
10799	082177364X	Finders Keepers	Fern Michaels	2002	Zebra Books
42019	0070465037	Finders Keepers	Barbara Nickolae	1989	McGraw-Hill Companies
58264	0688118461	Finders Keepers	Emily Rodda	1993	Harpercollins Juvenile Books
66678	1575663236	Finders Keepers	Fern Michaels	1998	Kensington Publishing Corporation
129037	1931696993	Finders Keepers	Linnea Sinclair	2001	NaN
134309	0156309505	Finders Keepers	Will	1989	Voyager Books
173473	0973146907	Finders Keepers	Sean M. Costello	2002	Red Tower Publications
195885	0061083909	Finders Keepers	Sharon Sala	2003	HarperTorch
211874	0373261160	Finders Keepers	Elizabeth Travis	1993	Worldwide Library

## 由图书作者检查以找到模式## 都有不同的出版商。这里没有线索
books.loc[(books.bookAuthor == 'Elaine Corvidae'),:]

	ISBN	bookTitle	bookAuthor	yearOfPublication	publisher
126762	1931696934	Winter's Orphans	Elaine Corvidae	2001	Novelbooks
128890	193169656X	Tyrant Moon	Elaine Corvidae	2002	NaN
129001	0759901880	Wolfkin	Elaine Corvidae	2001	Hard Shell Word Factory

## 由图书作者检查以找到模式
books.loc[(books.bookAuthor == 'Linnea Sinclair'),:]

	ISBN	bookTitle	bookAuthor	yearOfPublication	publisher
129037	1931696993	Finders Keepers	Linnea Sinclair	2001	NaN

## 因为没有什么共同的东西可以推断出NaNs的发布者，将它们替换为“other”
books.loc[(books.ISBN == '193169656X'),'publisher'] = 'other'
books.loc[(books.ISBN == '1931696993'),'publisher'] = 'other'

## 二、用户数据集
print (users.shape)
users.head()

(278858, 3)

	userID	Location	Age
0	1	nyc, new york, usa	NaN
1	2	stockton, california, usa	18.0
2	3	moscow, yukon territory, russia	NaN
3	4	porto, v.n.gaia, portugal	17.0
4	5	farnborough, hants, united kingdom	NaN

users.dtypes

userID        int64
Location     object
Age         float64
dtype: object

users.userID.values

array([     1,      2,      3, ..., 278856, 278857, 278858], dtype=int64)

## Age 
sorted(users.Age.unique())

[nan,0.0,1.0,2.0,3.0,4.0,5.0,6.0,7.0,8.0,9.0,10.0,11.0,12.0,13.0,14.0,15.0,16.0,17.0,18.0,19.0,20.0,21.0,22.0,23.0,24.0,25.0,26.0,27.0,28.0,29.0,30.0,31.0,32.0,33.0,34.0,35.0,36.0,37.0,38.0,39.0,40.0,41.0,42.0,43.0,44.0,45.0,46.0,47.0,48.0,49.0,50.0,51.0,52.0,53.0,54.0,55.0,56.0,57.0,58.0,59.0,60.0,61.0,62.0,63.0,64.0,65.0,66.0,67.0,68.0,69.0,70.0,71.0,72.0,73.0,74.0,75.0,76.0,77.0,78.0,79.0,80.0,81.0,82.0,83.0,84.0,85.0,86.0,87.0,88.0,89.0,90.0,91.0,92.0,93.0,94.0,95.0,96.0,97.0,98.0,99.0,100.0,101.0,102.0,103.0,104.0,105.0,106.0,107.0,108.0,109.0,110.0,111.0,113.0,114.0,115.0,116.0,118.0,119.0,123.0,124.0,127.0,128.0,132.0,133.0,136.0,137.0,138.0,140.0,141.0,143.0,146.0,147.0,148.0,151.0,152.0,156.0,157.0,159.0,162.0,168.0,172.0,175.0,183.0,186.0,189.0,199.0,200.0,201.0,204.0,207.0,208.0,209.0,210.0,212.0,219.0,220.0,223.0,226.0,228.0,229.0,230.0,231.0,237.0,239.0,244.0]

## 年龄栏有一些无效的条目，比如nan，0和非常高的值，比如100和以上
users.loc[(users.Age > 90) | (users.Age < 5), 'Age'] = np.nan

## 用平均值代替NaN
## 将数据类型设置为int
users.Age = users.Age.fillna(users.Age.mean())
users.Age = users.Age.astype(np.int32)

sorted(users.Age.unique())

[5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90]

## 三、评级数据集
ratings.shape

(1149780, 3)

## 如果每个用户对每个条目进行评级，那么评级数据集将有nusers * nbooks条目，这表明数据集非常稀疏。
n_users = users.shape[0]
n_books = books.shape[0]
print (n_users * n_books)

75670906880

ratings.head(5)

	userID	ISBN	bookRating
0	276725	034545104X	0
1	276726	0155061224	5
2	276727	0446520802	0
3	276729	052165615X	3
4	276729	0521795028	6

ratings.bookRating.unique()

array([ 0,  5,  3,  6,  8,  7, 10,  9,  4,  1,  2], dtype=int64)

ratings_new = ratings[ratings.ISBN.isin(books.ISBN)]

print (ratings.shape)
print (ratings_new.shape)

(1149780, 3)
(1031136, 3)

## 没有新用户添加，因此我们将使用高于数据集的新用户（1031136，3）
print ("number of users: " + str(n_users))
print ("number of books: " + str(n_books))

number of users: 278858
number of books: 271360

sparsity=1.0-len(ratings_new)/float(n_users*n_books)
print ('图书交叉数据集的稀疏级别是 ' +  str(sparsity*100) + ' %')

图书交叉数据集的稀疏级别是 99.99863734155898 %

ratings.bookRating.unique()

array([ 0,  5,  3,  6,  8,  7, 10,  9,  4,  1,  2], dtype=int64)

ratings_explicit = ratings_new[ratings_new.bookRating != 0]
ratings_implicit = ratings_new[ratings_new.bookRating == 0]

print (ratings_new.shape)
print( ratings_explicit.shape)
print (ratings_implicit.shape)

(1031136, 3)
(383842, 3)
(647294, 3)

## 统计
sns.countplot(data=ratings_explicit , x='bookRating')
plt.show()

在这里插入图片描述

## 基于简单流行度的推荐系统
ratings_count = pd.DataFrame(ratings_explicit.groupby(['ISBN'])['bookRating'].sum())
top10 = ratings_count.sort_values('bookRating', ascending = False).head(10)
print ("推荐下列书籍")
top10.merge(books, left_index = True, right_on = 'ISBN')

推荐下列书籍

	bookRating	ISBN	bookTitle	bookAuthor	yearOfPublication	publisher
408	5787	0316666343	The Lovely Bones: A Novel	Alice Sebold	2002	Little, Brown
748	4108	0385504209	The Da Vinci Code	Dan Brown	2003	Doubleday
522	3134	0312195516	The Red Tent (Bestselling Backlist)	Anita Diamant	1998	Picador USA
2143	2798	059035342X	Harry Potter and the Sorcerer's Stone (Harry P...	J. K. Rowling	1999	Arthur A. Levine Books
356	2595	0142001740	The Secret Life of Bees	Sue Monk Kidd	2003	Penguin Books
26	2551	0971880107	Wild Animus	Rich Shapero	2004	Too Far
1105	2524	0060928336	Divine Secrets of the Ya-Ya Sisterhood: A Novel	Rebecca Wells	1997	Perennial
706	2402	0446672211	Where the Heart Is (Oprah's Book Club (Paperba...	Billie Letts	1998	Warner Books
231	2219	0452282152	Girl with a Pearl Earring	Tracy Chevalier	2001	Plume Books
118	2179	0671027360	Angels & Demons	Dan Brown	2001	Pocket Star

users_exp_ratings = users[users.userID.isin(ratings_explicit.userID)]
users_imp_ratings = users[users.userID.isin(ratings_implicit.userID)]

print (users.shape)
print (users_exp_ratings.shape)
print (users_imp_ratings.shape)

(278858, 3)
(68091, 3)
(52451, 3)

## 基于协同过滤的推荐系统
counts1 = ratings_explicit['userID'].value_counts()
ratings_explicit = ratings_explicit[ratings_explicit['userID'].isin(counts1[counts1 >= 100].index)]
counts = ratings_explicit['bookRating'].value_counts()
ratings_explicit = ratings_explicit[ratings_explicit['bookRating'].isin(counts[counts >= 100].index)]

ratings_matrix = ratings_explicit.pivot(index='userID', columns='ISBN', values='bookRating')
userID = ratings_matrix.index
ISBN = ratings_matrix.columns
print(ratings_matrix.shape)
ratings_matrix.head()

(449, 66574)

ISBN	0000913154	0001046438	000104687X	0001047213	0001047973	000104799X	0001048082	0001053736	0001053744	0001055607	...	B000092Q0A	B00009EF82	B00009NDAN	B0000DYXID	B0000T6KHI	B0000VZEJQ	B0000X8HIE	B00013AX9E	B0001I1KOG	B000234N3A
userID
2033	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
2110	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
2276	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
4017	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
4385	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN

5 rows × 66574 columns

n_users = ratings_matrix.shape[0] #只考虑那些给出明确评级的用户
n_books = ratings_matrix.shape[1]
print (n_users, n_books)

449 66574

ratings_matrix.fillna(0, inplace = True)
ratings_matrix = ratings_matrix.astype(np.int32)

ratings_matrix.head(5)

ISBN	0000913154	0001046438	000104687X	0001047213	0001047973	000104799X	0001048082	0001053736	0001053744	0001055607	...	B000092Q0A	B00009EF82	B00009NDAN	B0000DYXID	B0000T6KHI	B0000VZEJQ	B0000X8HIE	B00013AX9E	B0001I1KOG	B000234N3A
userID
2033	0	0	0	0	0	0	0	0	0	0	...	0	0	0	0	0	0	0	0	0	0
2110	0	0	0	0	0	0	0	0	0	0	...	0	0	0	0	0	0	0	0	0	0
2276	0	0	0	0	0	0	0	0	0	0	...	0	0	0	0	0	0	0	0	0	0
4017	0	0	0	0	0	0	0	0	0	0	...	0	0	0	0	0	0	0	0	0	0
4385	0	0	0	0	0	0	0	0	0	0	...	0	0	0	0	0	0	0	0	0	0

5 rows × 66574 columns

sparsity=1.0-len(ratings_explicit)/float(users_exp_ratings.shape[0]*n_books)
print ('图书交叉数据集的稀疏级别是 ' +  str(sparsity*100) + ' %')

图书交叉数据集的稀疏级别是 99.99772184106935 %

## 基于用户的协同过滤
global metric,k
k=10
metric='cosine'
def findksimilarusers(user_id, ratings, metric = metric, k=k):similarities=[]indices=[]model_knn = NearestNeighbors(metric = metric, algorithm = 'brute') model_knn.fit(ratings)loc = ratings.index.get_loc(user_id)distances, indices = model_knn.kneighbors(ratings.iloc[loc, :].values.reshape(1, -1), n_neighbors = k+1)similarities = 1-distances.flatten()return similarities,indices

def predict_userbased(user_id, item_id, ratings, metric = metric, k=k):prediction=0user_loc = ratings.index.get_loc(user_id)item_loc = ratings.columns.get_loc(item_id)similarities, indices=findksimilarusers(user_id, ratings,metric, k) #similar users based on cosine similaritymean_rating = ratings.iloc[user_loc,:].mean() #to adjust for zero based indexingsum_wt = np.sum(similarities)-1product=1wtd_sum = 0 for i in range(0, len(indices.flatten())):if indices.flatten()[i] == user_loc:continue;else: ratings_diff = ratings.iloc[indices.flatten()[i],item_loc]-np.mean(ratings.iloc[indices.flatten()[i],:])product = ratings_diff * (similarities[i])wtd_sum = wtd_sum + product#在非常稀疏的数据集的情况下，使用基于协作的方法的相关度量可能会给出负面的评价#在这里的处理如下if prediction <= 0:prediction = 1   elif prediction >10:prediction = 10prediction = int(round(mean_rating + (wtd_sum/sum_wt)))print ('用户预测等级 {0} -> item {1}: {2}'.format(user_id,item_id,prediction))return prediction

## 测试
predict_userbased(11676,'0001056107',ratings_matrix)

用户预测等级 11676 -> item 0001056107: 22

## 基于项目的协同过滤
def findksimilaritems(item_id, ratings, metric=metric, k=k):similarities=[]indices=[]ratings=ratings.Tloc = ratings.index.get_loc(item_id)model_knn = NearestNeighbors(metric = metric, algorithm = 'brute')model_knn.fit(ratings)distances, indices = model_knn.kneighbors(ratings.iloc[loc, :].values.reshape(1, -1), n_neighbors = k+1)similarities = 1-distances.flatten()return similarities,indices

def predict_itembased(user_id, item_id, ratings, metric = metric, k=k):prediction= wtd_sum =0user_loc = ratings.index.get_loc(user_id)item_loc = ratings.columns.get_loc(item_id)similarities, indices=findksimilaritems(item_id, ratings) #similar users based on correlation coefficientssum_wt = np.sum(similarities)-1product=1for i in range(0, len(indices.flatten())):if indices.flatten()[i] == item_loc:continue;else:product = ratings.iloc[user_loc,indices.flatten()[i]] * (similarities[i])wtd_sum = wtd_sum + product                              prediction = int(round(wtd_sum/sum_wt))#在非常稀疏的数据集的情况下，使用基于协作的方法的相关度量可能会给出负面的评价#在这里处理的是下面的//代码，没有下面的代码片段，下面的代码片段是为了避免负面影响#在使用相关度规时，可能会出现非常稀疏的数据集的预测if prediction <= 0:prediction = 1   elif prediction >10:prediction = 10print ('用户预测等级 {0} -> item {1}: {2}'.format(user_id,item_id,prediction)    )  return prediction

## 测试
prediction = predict_itembased(11676,'0001056107',ratings_matrix)

用户预测等级 11676 -> item 0001056107: 1

查看全文

http://www.lryc.cn/news/144094.html

在腾讯云服务器OpenCLoudOS系统中安装svn（有图详解)

C语言日常刷题5

【LeetCode-中等题】73. 矩阵置零

本地部署 FastGPT

软件工程(十八) 行为型设计模式(四)

Socket通信与WebSocket协议

新KG视点 | Jeff Pan、陈矫彦等——大语言模型与知识图谱的机遇与挑战

详解过滤器Filter和拦截器Interceptor的区别和联系

List常用的操作

Android studio APK切换多个摄像头(Camera2)

ChatGPT 对教育的影响，AI 如何颠覆传统教育

Spring(九）声明式事务

java中用HSSFWorkbook生成xls格式的excel(亲测)

做平面设计一般电脑可以吗优漫动游

设计模式备忘录+命令模式实现Word撤销恢复操作

Linux centos7 bash编程小训练

创作2周年纪念日-特别篇

【UE5】用法简介-使用MAWI高精度树林资产的地形材质与添加风雪效果

兼容AD210 车规级高精度隔离放大器：ISO EM210

R语言常用数组函数

前端开发之Element Plus的分页组件el-pagination显示英文转变为中文

基于Java+SpringBoot+Vue前后端分离社区医院管理系统设计和实现

浅谈单例模式在游戏开发中的应用

Stable Diffusion WebUI 整合包

什么是 RESTful API

如何搭建关键字驱动自动化测试框架？

WPF实战项目十二（API篇）：配置AutoMapper

Linux 内核模块加载过程之重定位

Flink流批一体计算（19）：PyFlink DataStream API之State

相关文章：