python 中的错误：索引 0 超出尺寸为 0 的轴 0 的范围

Question

Today i want to learn about how to code a content based filtering in python, and so i search some code and i apply it. I have a simple dataset contains a hotel dataset, with the name, address, and description. After i tried the code, its said index 0 is out of bounds for axis 0 with size 0 at the end of the code. Here's the code:

import pandas as pd
import numpy as np
from sklearn.metrics.pairwise import cosine_similarity
from sklearn.feature_extraction.text import TfidfVectorizer
from nltk.corpus import stopwords
import re
import random

data = pd.read_csv('hotel.csv')
data.head()

the output: nama alamat deskripsi 0 Capital O 253 Topas Galeria Hotel Jl. Dr. Djundjunan No. 153, 40173 Bandung, Ind... Berjarak 10 menit berkendara dari Bandara Inte... 1 Sheraton Bandung Hotel & Towers Jl. Ir H Juanda 390, 40135 Bandung, Indonesia Sheraton Hotel & Towers menawarkan akomodasi b... 2 OYO 794 Ln 9 Bandung Residence Jalan Lemahnendeut No 9, Sukajadi, 40164 Bandu... Berlokasi nyaman di Sukajadi, Bandung, OYO 794... 3 OYO 226 LJ hotel Jl. Malabar No.2, Malabar, Lengkong, Dago, Asi... OYO 226 LJ hotel di Bandung, Jawa Barat, tepat... 4 OYO 230 Maleo Residence JI. Dangeur Indah II No. 15, Sukagalih, Sukaja... OYO 230 Maleo Residence menawarkan akomodasi b...

data.describe()
data.info()

the output:

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 90 entries, 0 to 89
Data columns (total 3 columns):
 #   Column     Non-Null Count  Dtype 
---  ------     --------------  ----- 
 0   nama       90 non-null     object
 1   alamat     90 non-null     object
 2   deskripsi  90 non-null     object
dtypes: object(3)
memory usage: 2.2+ KB

---

clean_spcl = re.compile('[/(){}\[\]\|@,;]')
clean_symbol = re.compile('[^0-9a-z #+_]')
stopworda = set(stopwords.words('english'))

def clean_text(text):
    text = text.lower() 
    text = clean_spcl.sub(' ', text)
    text = clean_symbol.sub('', text)
    text = ' '.join(word for word in text.split() if word not in stopworda) # hapus stopword dari kolom deskripsi
    return text
  
data['deskripsi_new'] = data['deskripsi'].apply(clean_text)

def pt_desc(index):
    example = data[data.index == index][['deskripsi_new', 'nama', 'alamat']].values[0]
    if len(example) > 0:
        print(example[0])
        print('Nama:', example[1])
        print('Alamat:', example[2])   

data.set_index('nama', inplace=True)
tf = TfidfVectorizer(analyzer='word', ngram_range=(1, 3), min_df=0, stop_words='english')
tfidf_matrix = tf.fit_transform(data['deskripsi_new'])
cos_sim = cosine_similarity(tfidf_matrix, tfidf_matrix)
cos_sim

the output:

array([[1.        , 0.07106689, 0.03075961, ..., 0.07474134, 0.0732575 ,
        0.01680878],
       [0.07106689, 1.        , 0.03508807, ..., 0.05947269, 0.08705608,
        0.01986701],
       [0.03075961, 0.03508807, 1.        , ..., 0.09113962, 0.05879732,
        0.06808138],
       ...,
       [0.07474134, 0.05947269, 0.09113962, ..., 1.        , 0.06321301,
        0.02205802],
       [0.0732575 , 0.08705608, 0.05879732, ..., 0.06321301, 1.        ,
        0.02245328],
       [0.01680878, 0.01986701, 0.06808138, ..., 0.02205802, 0.02245328,
        1.        ]])

indices = pd.Series(data.index)
indices[:50]

def rekomendasi(nama, cos_sim = cos_sim):
    
    rec = []
    
    idx = indices[indices == nama].index[0]

    score_series = pd.Series(cos_sim[idx]).sort_values(ascending = False)

    top_10_indexes = list(score_series.iloc[1:11].index)
    
    for i in top_10_indexes:
        recommended_news.append(list(data.index)[i])
        
    return rec

rekomendasi('Hotel') # and when i reach here, the error said 'index 0 is out of bounds for axis 0 with size 0'

what went wrong here?

Answer 1

From what I understand you are trying to build a kind of search engine, which given a search vector will return the 10 best matching results.

If this is the case, you'll need to modify your rekomendasi function so that it will:

• process the input query vector
• compute the similarity scores with the corpus (corpus here mean the list of your hotel descriptions)
• return the 10 items with highest similarity scores

I've modified your code to do that:

import pandas as pd
import numpy as np
from sklearn.metrics.pairwise import cosine_similarity
from sklearn.feature_extraction.text import TfidfVectorizer
from nltk.corpus import stopwords
import re
import random

data = pd.read_csv('../../../../Downloads/hotel.csv')

clean_spcl = re.compile('[/(){}\[\]\|@,;]')
clean_symbol = re.compile('[^0-9a-z #+_]')
stopworda = set(stopwords.words('english'))

def clean_text(text):
    text = text.lower() 
    text = clean_spcl.sub(' ', text)
    text = clean_symbol.sub('', text)
    text = ' '.join(word for word in text.split() if word not in stopworda) # hapus stopword dari kolom deskripsi
    return text
  
data['deskripsi_new'] = data['deskripsi'].apply(clean_text)

def pt_desc(index):
    example = data[data.index == index][['deskripsi_new', 'nama', 'alamat']].values[0]
    if len(example) > 0:
        print(example[0])
        print('Nama:', example[1])
        print('Alamat:', example[2])   

data.set_index('nama', inplace=True)
tf = TfidfVectorizer(analyzer='word', ngram_range=(1, 3), min_df=0, stop_words='english')
tfidf_matrix = tf.fit_transform(data['deskripsi_new'])
cos_sim = cosine_similarity(tfidf_matrix, tfidf_matrix)
cos_sim

def rekomendasi(nama, cos_sim=cos_sim):
    
    # you first need to preprocess the given query text (i.e. nama) and transform it in tf=idf vector
    nama = clean_text(nama)
    nama_vector = tf.transform([nama])
    
    # Next we compute similarity scores between the query text (nama) and the corpus (tfidf_matrix)
    similarity_scores = cosine_similarity(nama_vector, tfidf_matrix).squeeze()
    top_10_indices = similarity_scores.argsort()[-10:][::-1]
    
    rec = data.index[top_10_indices].tolist()
    return rec

Example:

rekomendasi('Hotel')

['The Trans Luxury Hotel Bandung', 'M Premiere Hotel Dago Bandung', 'Mutiara Hotel', 'éL Hotel Royale Bandung', 'The Jayakarta Suites Bandung, Hotel & Spa', 'Hotel Cemerlang', "OYO 167 Dago's Hill Hotel", "OYO 167 Dago's Hill Hotel", 'HARRIS Hotel & Conventions Ciumbuleuit – Bandung', 'Padma Hotel Bandung']