Overview

--Acquired information on Japanese infected persons of the new coronavirus (COVID-19) --Morphological analysis with mecab --Visualize feature words with wordcloud

reference

Information on infected persons of the new coronavirus (COVID-19)

https://qiita.com/ryo-ma/items/20db8cd20f1086838249 --Data that seems to be infected person information: https://raw.githubusercontent.com/swsoyee/2019-ncov-japan/master/Data/resultSignateDetail.csv

config

config.py

import re
import os

### MeCab
POS_LIST = [10, 11, 31, 32, 34]
POS_LIST.extend(list(range(36,50)))
POS_LIST.extend([59, 60, 62, 67])
STOP_WORDS = ["To do", "Absent", "Become", "Already", "Shiyo", "Can", "Became", "Ku", "Finally", "is there", "May", "think", "today", "It", "this", "that", "which one", "Which", "NULL", "To be", "Nari", "Ah", "Canる", "I"]
RE_ALPHABET = re.compile("^[0-9a-zA-Z０-９ .,*<>]+$") # alphabet, number, space, comma or dot
current_dir = os.getcwd()
OUTPUT_PNG_FILE = os.path.join(current_dir, "wordcloud.png ")

Preprocessing

(Omitted)

Morphological analysis

import MeCab
from os import path
from wordcloud import WordCloud
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import re

def create_mecab_list(text_list):
	mecab_list = []
	mecab = MeCab.Tagger("-Ochasen -d /usr/local/lib/mecab/dic/mecab-ipadic-neologd") # MacOS
	mecab.parse("")
	# encoding = text.encode('utf-8')
	for text in text_list:
		node = mecab.parseToNode(text)
		while node:
			# [Part of speech,Part of speech細分類1,Part of speech細分類2,Part of speech細分類3,Inflected form,Utilization type,Prototype,reading,pronunciation]
			#Busy adjective,Independence,*,*,Adjective, Idan,Continuous connection,busy,Isogasiku,Isogasiku
			morpheme = node.feature.split(",")[6]
			if RE_ALPHABET.match(morpheme):
				node = node.next
				continue
			if morpheme in STOP_WORDS:
				node = node.next
				continue
			if len(morpheme) > 1:
				if node.posid in POS_LIST:
					mecab_list.append(morpheme)
			node = node.next
	return mecab_list

wordcloud

import MeCab
from os import path
from wordcloud import WordCloud
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import re

def create_wordcloud(morphemes):
	# fpath = "/usr/share/fonts/truetype/takao-gothic/TakaoPGothic.ttf" # Ubuntu
	fpath = "/System/Library/Fonts/Hiragino Maru Go ProN W4.ttc" # Mac OS X
	wordcloud = WordCloud(
		background_color="whitesmoke",
		collocations=False,
		stopwords=set(STOP_WORDS),
		max_font_size=80,
		relative_scaling=.5,
		width=800,
		height=500,
		font_path=fpath
		).generate(morphemes)
	plt.figure()
	plt.imshow(wordcloud)
	plt.axis("off")
	wordcloud.to_file(OUTPUT_PNG_FILE)

result

wordcloud (5).png

――There are more infected people in "male" than in "female" -→ "Male" has a larger font size than "Female" ――There are many other than "20's" --"Mask" is important ...

Other new corona related information

--Ministry of Health, Labor and Welfare --Q & A about the new coronavirus: - https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/kenkou_iryou/dengue_fever_qa_00001.html ――Guidelines for consultation and consultation regarding new coronavirus infections - https://www.mhlw.go.jp/content/10900000/000596905.pdf

Tokyo --New coronavirus infection control site - https://stopcovid19.metro.tokyo.lg.jp/ --Visualization of infected people in the world
- https://github.com/sorxrob/covid-3d --Globe https://covid3d.live --Dashboard made by Uber engineers --PC version: https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6 --Mobile version: https://www.arcgis.com/apps/opsdashboard/index.html#/85320e2ea5424dfaaa75ae62e5c06e61