My project starts with this sentence !
Today, whether it is in the media
(See Video France 3 on one of the actions taken in the company for
the well-being of employees), in the so-called traditional press with special sections on the
subject such as for the newspaper
Le Monde
but also on TEDTalks videos on
well-being at work
or in less traditional media such as
Welcome to the Jungle , well-being at work is a real current topic. However, it is still
difficult to find out what we really like, whether it is before
starting studies, during studies, after studies or during our
professional career.
I decided to create a software called "Happy at work" which aims to help company employees to better understand what they like and dislike. I believe that Artificial Intelligence can increase performance and well-being at work. Indeed, many people speak about that such as PwC as we can see here :
But also there are lot of report and documentary about that such as an explanation of "How A.I. is helping boost workplace satisfaction" :
Source (video)
You'll say to me "But if you do a task and you don't like to do it,
you know it, and conversely, when I'm angry about a job, I know it
too...".
Yes, BUT after a while, work becomes a certain routine. So it can be
difficult to grasp what we really like to do in our work, which on
the contrary, we hate to do.
What the software also offers is to have a long-term vision and not
a vision at a precise moment and to be able to know more about
yourself over time.
First of all, my project would be used to recognize the person on
his workstation and would carry out a check to prove that it is him
through facial recognition.
But above all, my software would allow employees to get to know
themselves better, to better understand their interests, to better
understand their emotions, to increase their performance and to
promote their well-being at work.
This data will be purely for the employee: he will be able to use it
in interviews but that's not the point, it's above all a help for
him.
The employee will be able to see at what moment of the day he is
happy in front of his work and conversely at what moment he is
unhappy/stressed/angry.
Also, to see the moments when he is most tired. This application
would allow him to correct this if he wishes.
First of all, when the employee logs on to his or her computer, the
software must recognize his or her face.
We have to use
ml5.js where we find
an API to recognise people's faces. This tool is called FaceApi.Net.
Face-Api
allows to access face and face landmark detection.
You can see bellow photos in order to better understand the model
(photos are from this website) : For more information, click
here
:
It is also necessary to use the Face Expression Recognition Model to
be able to see the emotions of the employees and to be able to know if
they are happy, sad, angry.
You can see bellow photos in order to better understand the model
(photos are from this website) : For more information, click
here
:
For more details here is an example of what I would like to have to make my application work. This example comes from the example of ml5.js. posted on Github. We can see below the step for the model :
- On
Github
there are lot of models and explanation !
- A tutorial that explains how to use
Face Detection
!
- An article to learn more about
Face-Api
and how to use it !
Then, in order to my project become optimum, I would have to create a
curve with the data so that it transmits the data in a table and
people can see according to the time what they were doing when they
were stressed, sad or happy.
But How it will work ?
The person could agree to be filmed all day and then take some time
to view the images. When the person is really angry, an alarm signal
could be given by the computer and the person could see directly
through the moments when they felt stress, anger or happy.
The linking of this data (between what the video is filming with the
person's moods at a certain time and in a certain way and the
person's TO DO list and what he or she is doing at a given time)
should be developed into a graph.
To do this, I would like to call on an Institute of Statistics, a
school of computer science and data study that will be at the
forefront in this field. He/she will thus be able to create a
correlation between the two in order to be able to assess the
well-being and satisfaction of employees according to the tasks
performed.
To find out more about this and about face-api you can click
here
.