Cloud AI to guess blood type from face photo

I found an interesting article "challenge to make up Raspberry Pi AI to guess blood type from face photo".
However, the result is rather disappointing -- its accuracy rate is as low as 27.8%, so it does not seem to "guess". The source is a Japanese magazine "Interface" April 2017 issue by CQ Publishing (Japan). Anyway, this is a joyful experiment.
Unfortunately, this article is written in Japanese without English title, I will introduce the overview.

Source: the company's site.
(Note: Deep Leaning -> Deep Learning)

Image data are divided into 128 x 128 areas (pixels), so the resolution is comparatively low (e. g. that of HDTV is 1920 x 1080 pixels).
Disappointed by the poor result, researchers focused on various features such as nose, correcting face images to the front direction, but things did not get better.
I think it is too early to conclude that "there is no correlation between face photos and blood types" in the experiment alone ...

<Reference data>

[Image data] Convert 6000 photos to grayscale 128 x 128 pixels
[Equipments for data learning] i7-4770 / 3.4 GHz / 16 GB (DDR 3) + Amazon cloud GPU (AWS EC 2 p 2. Xlarge parg) + machine learning library dlib
[Equipments for blood type determination] Raspberry Pi 3 / PiCamera (960 × 540 pixels)
[Processing time] About 20 seconds for face detection, about 30 seconds for blood type determination
[Results] The accuracy rate of test data is as low as 27.8%! Thereafter, various measures are taken, but accuracy rates showed little improvement. Researchers finally concluded that "there is no correlation between face photos and blood types."

<Source>

Hitoaki NAKAMURA, Satoru IWASADA (Bee Co., Ltd, Osaka Japan)
Cloud is good for studying image deep learning! A challenge to make up Raspberry Pi AI to guess blood type from face photo (Featured specials for beginners! RasPi x cloud AI: GPUs from 100 yen!)
Interface magazine (April 2017 issue) by CQ Publishing Co., Limited (Tokyo Japan).

中村 仁昭 , 岩貞 智
画像ディープ・ラーニングの学習はクラウドが良し! 顔写真から血液型を当てるラズパイ人工知能に挑戦してみた (注目特集 ビギナ向け! ラズパイ×クラウド人工知能 : GPUも100円から!)  [in Japanese]

Research and development activities are on the company's site. (in Japanese)

Blood Types of the Rich and Famous

According to a website Plasma Donation, blood types of the rich and famous peole are ...

<Type O>
Musicians John Lennon, Elvis Presley, Liam & Noel Gallagher, Crystal Kay; actor Paul Newman; and world leaders Queen Elizabeth II, Gerald Ford, Ronald Reagan and Mikhail Gorbachev.

<Type A>
Former U.S. Presidents George Bush Senior, Lyndon B. Johnson and Richard Nixon; musicians Ringo Starr, Britney Spears, Rick James and Maki Nomiya; actors Alan Alda and Jet Li; German politician Adolf Hitler; and Japanese novelist Soseki Natsume.

<Type B>
Actors Jack Nicholson, Leonardo DiCaprio and Mia Farrow; musicians Paul McCartney and Luciano Pavarotti; Japanese film director Akira Kurosawa; and football player Vince Young.

<Type AB>
Actors Jackie Chan and Marilyn Monroe; musician Mick Jagger; former U.S. President John F. Kennedy; inventor Thomas Edison; and mixed martial artist Bob Sapp.

I have made my English Mobile Site

I have made my English mobile site using Jimdo.
https://abofan-global.jimdo.com/

It is not necessary to handmade from scratch like the old days.
Now using CMS like Jimdo or Wix, and Google translation, making an English website has became much easier.
But, it is still a tough work for me who is not an English native.

My English mobile site is for foreigners, the atmosphere and contents are a little bit different from the original Japanese version.
# In short, expressions are straightforward and pretentious with a lot of data.

It is hard to make everything original, so I have diverted a lot of old contents.
Sadly, my English is so terrible in old contents, especially.
I did quite a bit of reworking ...

Now that I noticed Koreans and Chinese are more enthusiastic about blood-type research. If it goes on like this, they might overtake Japan like the the economy ...

Mr. Fred Wong & Eugenia Wan published a book "Bloody AI Alchemist"

It seems to be quite full-fledged contents and I am expecting very much.
Aauthors are working for an investment bank, and seem to be strong for AI.
Because written in Enlish, it will take time to read, because I am Japanese ...
The authors' site is as follows.
http://www.bloodtype.ai/english/
They said that they are currently looking for collaborators.
The address of their Facebook is as follows.
https://www.facebook.com/bloodtypeai/

Three Research Reports of the Middle East and One Indian

I found three research reports of the Middle East for the first time (luckyly, it is written in English).
Perhaps, famous Korean cartoon "RealCrazyMan's Blood Types Comic" was translated into Persian or Arabic ....
#It is exactly what I expect, that the results do not match the other reports.

The reference of the first paper is Mr. Alexander Besher's "You are Your Blood Type," so I am keenly aware of the necessity of transmitting English information.
By the way, these Iranian research reports are not registered in English-language database,  somehow.
Is there any reason?

<No.1>
Fatemeh Beheshtian, Roghayeh Hashemi and Zolfaghar Rashidi
The Five Personality Factors over the Students with Four Blood Types
Journal of Applied Environmental and Biological Sciences, 2015, 5(8):45-49 1
160 subjects, NEO PI-R -> diffrence in openness to experience
https://www.textroad.com/pdf/JAEBS/J.%20Appl.%20Environ.%20Biol.%20Sci.,%205(8)45-49,%202015.pdf

<No.2>
Mohammad Sharifi, Hamza Ahmadian and Ali Jalali
The relationship between the big five personality factors with blood types in Iranian university students
Journal of Chemical and Pharmaceutical Research, 2015, 7(5):233-240
400 subjects, NEO PI -> diffrences in openness and extraversion
http://www.jocpr.com/articles/the-relationship-between-the-big-five-personality-factors-with-blood-types-iniranian-university-students.pdf

<No.3>
Zirak Morandlou Hossein
The Relationship between Students’Personality Traits and Their Blood Types
Journal of Health and Devalopment 2012, 1(3):221-226
number of subjects not specified NEO-FFI -> no diffrence found
http://en.journals.sid.ir/ViewPaper.aspx?ID=299581

Also I found an Indian research report. This is not registered in English-language database, too.

Amreen Nahida, Nandini Chatterjee
A study on relationship between blood group and personality
International Journal of Home Science 2016; 2(1): 239-243
100 subjects -> no diffrence found
Eysenck’s Personality Questionnaire-Revised (EPQ-R) is used
http://www.homesciencejournal.com/archives/2016/vol2issue1/PartD/2-1-54.pdf