Motivation for development

New Year's cards sent every year ... When I was organizing my room, I got a lot of New Year's cards from my seniors and juniors. I wondered if this could be useful.

Then I came up with the idea that I could visualize myself as seen by others through the New Year's card. I wondered if so-called self-analysis could be done through New Year's cards.

Come to think of it, when I write to another person, I write with that person's impressions and episodes from last year. I wondered if this was the same for others.

You should be able to extract your impression of yourself by morphologically analyzing the New Year's card ... I wanted to make it a word cloud and visualize my impression of others.

How to make

Type in the text of the 1st New Year's card (data entry)

First, we need to collect the data to be analyzed, so we will summarize the contents of the New Year's card in Excel.

Like this, I entered it without the ridiculous greetings of Happy New Year and Kotoyoro. As much as possible, I tried to enter only words related to my impressions and episodes.

2 Collect data

Next, combine the entered Excel into one data

`python`


import xlrd

wb = xlrd.open_workbook('/nenga2020.xlsx')
sheet = wb.sheet_by_name('Sheet1')
col_values = sheet.col_values(0)
text=""
for i in col_values:
    text=text+i
print(text)

This means that the text contains all the text of the New Year's card.

3 Morphological analysis & word cloud creation

Finally, morphological analysis and word cloud creation are done from here.


import MeCab
import wordcloud, codecs

m = MeCab.Tagger("")
text = text.replace('\r', '')
parsed = m.parse(text)

splitted = ' '.join(
    [x.split('\t')[0] for x in parsed.splitlines()[:-1] if x.split('\t')[1].split(',')[0]  in ["noun","adjective","Adjectival noun"] ])

wordc = wordcloud.WordCloud(font_path='HGRGM.TTC',
                            background_color='white',
                            contour_color='steelblue',
                            contour_width=2).generate(splitted)
wordc.to_file('nenga2020.png')

This will solve the impression written on the New Year's card.

`python`


splitted = ' '.join(
    [x.split('\t')[0] for x in parsed.splitlines()[:-1] if x.split('\t')[1].split(',')[0]  in ["noun","adjective","Adjectival noun"] ])

By the way, the part of speech is narrowed down to nouns, adjectives, and adjective verbs. This is because the purpose is to extract my impression.

Summary

Whole code

import xlrd
import MeCab
import wordcloud, codecs


wb = xlrd.open_workbook('/nenga2020.xlsx')
sheet = wb.sheet_by_name('Sheet1')
col_values = sheet.col_values(0)
text=""
for i in col_values:
    text=text+i

m = MeCab.Tagger("")
text = text.replace('\r', '')
parsed = m.parse(text)

splitted = ' '.join(
    [x.split('\t')[0] for x in parsed.splitlines()[:-1] if x.split('\t')[1].split(',')[0]  in ["noun","adjective","Adjectival noun"] ])

wordc = wordcloud.WordCloud(font_path='HGRGM.TTC',
                            background_color='white',
                            contour_color='steelblue',
                            contour_width=2).generate(splitted)
wordc.to_file('nenga2020.png')

And here is the completed word cloud for self-analysis.