スクレイピング

code:R

# パッケージの読み込み

install.packages(c("rvest", "dplyr", "purrr", "stringr", "magrittr", "tidyr", "lubridate")) # ← 追加で tidyr, lubridate も必要

library(rvest)

library(dplyr)

library(purrr)

library(stringr)

library(magrittr)

library(tidyr)

library(lubridate)

# 川崎市（川崎区）のベースURL

# スクレイピング関数（1ページ分）

scrape_suumo_page <- function(url) {

message("アクセス中: ", url)

page <- tryCatch({

read_html(url)

}, error = function(e) {

warning("読み込み失敗: ", url)

return(tibble())

})

listings <- page %>% html_elements(".cassetteitem")

if (length(listings) == 0) {

warning("物件が見つかりませんでした: ", url)

return(tibble())

}

map_dfr(listings, function(listing) {

tibble(

name = listing %>% html_element(".cassetteitem_content-title") %>% html_text(trim = TRUE),

address = listing %>% html_element(".cassetteitem_detail-col1") %>% html_text(trim = TRUE),

access = listing %>% html_element(".cassetteitem_detail-col2") %>% html_text(trim = TRUE),

age = listing %>% html_element(".cassetteitem_detail-col3") %>% html_text(trim = TRUE),

rent = listing %>% html_element(".cassetteitem_price--rent") %>% html_text(trim = TRUE),

management_fee = listing %>% html_element(".cassetteitem_price--administration") %>% html_text(trim = TRUE),

deposit = listing %>% html_element(".cassetteitem_price--deposit") %>% html_text(trim = TRUE),

key_money = listing %>% html_element(".cassetteitem_price--gratuity") %>% html_text(trim = TRUE),

layout = listing %>% html_element(".cassetteitem_madori") %>% html_text(trim = TRUE),

area = listing %>% html_element(".cassetteitem_menseki") %>% html_text(trim = TRUE)

)

})

}

# 最大ページ数を取得してからURLを作る

get_max_page <- function(url) {

page <- read_html(url)

page %>%

html_elements(".pagination-parts a") %>%

html_text(trim = TRUE) %>%

as.integer() %>%

max(na.rm = TRUE)

}

max_page <- get_max_page(base_url)

page_urls <- paste0(base_url, "&page=", 1:max_page)

# 全ページ分まとめて取得

suumo_data <- map_dfr(page_urls, scrape_suumo_page)

# ---------- ここからデータ整形処理を追加 ----------

suumo_data <- suumo_data %>%

mutate(

# 家賃（"6.8万円"）→ 68000円に変換（"-" や "相談" 対応）

rent_value = ifelse(str_detect(rent, "^\\d+(\\.\\d+)?万円$"),

as.numeric(str_remove(rent, "万円")) * 10000,

NA_real_),

# 管理費（"2000円", "5000円", "-"）→ 数値（円）に変換

management_fee_value = ifelse(str_detect(management_fee, "^\\d+円$"),

as.numeric(str_remove(management_fee, "円")),

0), # "-"や空欄は0円と仮定

# 家賃＋管理費の合計

total_cost = rent_value + management_fee_value,

# 面積（"25.3m2"）を数値に変換

area = as.numeric(str_extract(area, "\\d+(\\.\\d+)?")),

# 徒歩分数（"歩3分", "歩10分", …）をすべて抽出して、その中の最小値

walk_from_station = str_extract_all(access, "歩\\d+分"),

walk_from_station = map_int(walk_from_station, function(x) {

mins <- as.integer(str_extract(x, "\\d+"))

if (length(mins) == 0 || all(is.na(mins))) return(NA_integer_)

min(mins, na.rm = TRUE)

}),

# 築年数（"築10年" → 10, "新築" → 0）

age_of_building = case_when(

str_detect(age, "新築") ~ 0L,

str_detect(age, "築\\d+年") ~ as.integer(str_extract(age, "\\d+")),

TRUE ~ NA_integer_

)

) %>%

# 駅名と路線名を分離（"〇〇線「△△」徒歩x分" に対応）

separate(access, sep = "「", into = c("line", "station"), fill = "right") %>%

separate(station, sep = "」", into = c("station", "drop"), fill = "right") %>%

select(-drop) %>% # ← 一時列を削除

# 不要な元データを削除（rent, management_fee, deposit, key_money）

select(-c(rent, management_fee, deposit, key_money))

# ---------- 整形ここまで ----------

# データ確認

print(head(suumo_data, 5))

print(nrow(suumo_data)) # 件数

# CSVに保存

write.csv(suumo_data, "suumo_kawasaki003.csv", row.names = FALSE, fileEncoding = "UTF-8")

cat("✅ suumo_kawasaki003.csv を保存しました！左のFilesタブからExport→Downloadできます。\n")

plot(suumo_data$area,suumo_data$total_cost)

plot(suumo_data$walk_from_station,suumo_data$total_cost)

plot(suumo_data$age_of_building,suumo_data$total_cost)

plot(suumo_data$area,suumo_data$total_cost)

m<-lm(suumo_data$total_cost~suumo_data$area)

abline(m)

https://gyazo.com/b6afe370637bdfa964a17758510cc6e2

plot(suumo_data$walk_from_station,suumo_data$total_cost)

n<-lm(suumo_data$total_cost~suumo_data$walk_from_station)

abline(n)

https://gyazo.com/1c97ae39bc6aa3382b6bce7b60e6ab3d

plot(suumo_data$age_of_building,suumo_data$total_cost)

l<-lm(suumo_data$total_cost~suumo_data$age_of_building)

abline(l)

https://gyazo.com/b45ccfa4e05e7f30a9118fe73bd1ec3f

model <- lm(total_cost ~ area + walk_from_station + age_of_building, data = suumo_data)

summary(model)

https://gyazo.com/b4b289c3dabb5203e40355f75fbc5a65

25m^2で駅から10分、年数が10年の物件は

39175.82+1970.60*25+531.09*10-571.42*10

88037.52円

ここよりも強気に設定するか、弱気に設定するか。