sentimentAnalysis/webscraping.R at master · zawropati/sentimentAnalysis · GitHub

1
2
3
4
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
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
#install.packages("tidyverse")
#install.packages("twitteR")
library(twitteR)
library(tm)
library(ggplot2)
library('tidyverse')
library('rvest')
library('dplyr')
library(DBI)
library(RMySQL)

#defining URLs abd reading the html from website
url2 <- "https://www.booking.com/reviews/nl/hotel/best-western-amsterdam-airport-hotel.en-gb.html?aid=356980;label=gog235jc-1FEgdyZXZpZXdzKIICOOgHSDNYA2ipAYgBAZgBCbgBB8gBDNgBAegBAfgBDYgCAagCA7gCvpnT7AXAAgE;sid=b075a00677b76feaaf6f19336ed7d1a3"
webpage2 <- read_html(url2)

url <- "https://www.booking.com/reviews/vn/hotel/au-lac.en-gb.html?aid=356980;label=gog235jc-1BEgdyZXZpZXdzKIICOOgHSDNYA2ipAYgBAZgBCbgBB8gBDNgBAegBAYgCAagCA7gCvpnT7AXAAgE;sid=b075a00677b76feaaf6f19336ed7d1a3"
webpage <- read_html(url)

#selecting html elements from website and converting them to text only
rank_data_html_neg <- html_nodes(webpage,'.review_neg')
rank_data_neg <- html_text(rank_data_html_neg)

rank_data_html_pos <- html_nodes(webpage,'.review_pos')
rank_data_pos <- html_text(rank_data_html_pos)

#creating data frames
df_neg <- data.frame("review" = rank_data_neg, "label" = "negative")
df_pos <- data.frame("review" = rank_data_pos, "label" = "positive")

#same for the other website
rank_data_html_neg2 <- html_nodes(webpage2,'.review_neg')
rank_data_neg2 <- html_text(rank_data_html_neg2)

rank_data_html_pos2 <- html_nodes(webpage2,'.review_pos')
rank_data_pos2 <- html_text(rank_data_html_pos2)

df_neg2 <- data.frame("review" = rank_data_neg2, "label" = "negative")
df_pos2 <- data.frame("review" = rank_data_pos2, "label" = "positive")

#binding df from both websites into one
finalPos <- bind_rows(df_pos,df_pos2)
finalNeg <- bind_rows(df_neg,df_neg2)

#small cleaning of webscraped reviews
final$review <- gsub("Nothing", "", final$review, perl = TRUE)

finalNeg <- finalNeg  %>%
  filter(str_remove(review, "[\n]"))

finalPos <- finalPos  %>%
  filter(str_remove(review, "[\n]"))

finalNeg <- finalNeg %>%
  filter(!str_detect(review, "nothing") )

finalPos <- finalPos %>%
  filter(!str_detect(review, "nothing") )

#insert the webscraped reviews to hotel_reviews db on MySQL
conMySQL <- dbConnect(MySQL(), dbname="hotel_reviews", user="root",password= "kurczak1", host="localhost")

dbWriteTable(conMySQL,name="positive_webscraped", value=finalPos, row.names=FALSE, overwrite=TRUE)
dbWriteTable(conMySQL,name="negative_webscraped", value=finalNeg, row.names=FALSE, overwrite=TRUE)


######
#barplots made from webscraped reviews
# build a corpus
myCorpus <- Corpus(VectorSource(finalPos$review))
# make all text lowercase
myCorpus <- tm_map(myCorpus, tolower)
# show first items
strwrap(myCorpus[1:15])
# remove punctuation
myCorpus <- tm_map(myCorpus, removePunctuation)
# show first items
strwrap(myCorpus[1:5])

# remove numbers
myCorpus <- tm_map(myCorpus, removeNumbers)
# show first items
strwrap(myCorpus[1:15])

removeURL <-function(x) gsub("http[[:alnum:]]*","",x)
myCorpus <-tm_map(myCorpus,removeURL)
# show first items
strwrap(myCorpus[11:15])

# remove stopwords
myStopwords <- stopwords('english')
myStopwords
#remove custom stopwords
myStopwords<- setdiff(myStopwords, c("hotel","room"))

myCorpus <- tm_map(myCorpus, removeWords, myStopwords)

# Building a document term matrix
myTdm <- TermDocumentMatrix(myCorpus, control = list(wordLengths=c(1,Inf)))
myTdm
rownames(myTdm)
# findFreqTerms(myTdm, lowfreq=10)
termFrequency <- rowSums(as.matrix(myTdm))

termFrequency <-subset(termFrequency, termFrequency>=15)

barplot(sort(termFrequency, decreasing=TRUE),las=2 )

#negative plot

myCorpusNeg <- Corpus(VectorSource(finalNeg$review))
# make all text lowercase
myCorpusNeg <- tm_map(myCorpusNeg, tolower)
# show first items
strwrap(myCorpusNeg[1:15])
# remove punctuation
myCorpusNeg <- tm_map(myCorpusNeg, removePunctuation)

# remove numbers
myCorpusNeg <- tm_map(myCorpusNeg, removeNumbers)

myCorpusNeg <-tm_map(myCorpusNeg,removeURL)

myCorpusNeg <- tm_map(myCorpusNeg, removeWords, myStopwords)
# show first items
strwrap(myCorpus[11:15])

# Building a document term matrix
myTdmNeg <- TermDocumentMatrix(myCorpusNeg, control = list(wordLengths=c(1,Inf)))
myTdmNeg
rownames(myTdmNeg)
# findFreqTerms(myTdm, lowfreq=10)
termFrequencyNeg <- rowSums(as.matrix(myTdmNeg))

termFrequencyNeg <-subset(termFrequencyNeg, termFrequencyNeg>=15)

barplot(sort(termFrequencyNeg, decreasing=TRUE),las=2 )