With the power of new technology, machines can learn about us and our world better than humans can.
Using the power of new technologies in artificial intelligence and machine learning, we can create a fun website where you can simply draw a letter or a couple words in the browser and the computer can determine your state of mind and your personality.
To enable the computer to learn well using these cutting edge algorithms, lots of handwriting data is required to enable accurate predictions and adequate cloud storage is required so the computer can run these computationally intensive algorithms at a fast enough speed. The data will match handwriting to data such as (are you usually on time, are you more introverted or extroverted, do you like tradition, are you imaginative, are you ambitous,etc.) and states of mind like are you(happy, worried, excited, nervous, etc.). (The more money donated equates to more features and more accurate prediction).
This website can be used to have fun with friends or you can use it to alter your personality. Studies have shown handwriting therapy is one of the best ways to alter someone's personality and with the power of new technology you can get constant feedback and more accurate "diagnosis" than traditional handwriting analysts.
If people using this website, start to widely agree that the predictions are accurate then it could have enormous applications in society. We could start using handwriting to find out if our children are feeling sad. People could use handwriting to discover personality traits they did not know they had. We could get a better understanding of the personality of the leaders of our society(Honest, devious, greedy, extroverted, introverted). People, governments, and employers could better understand what jobs and roles in society individuals are best suited for.
Here I created a demo of a simplistic version of the website using the only handwriting data of this kind that is online.
With funding, the website can be spruced, made public, and expanded to include a broader array of information. We could predict many of your personality traits and more about your state of mind instead of just predicting whether or not you are worried and only using an m.
Risks and challenges
One of the main risks is that the models might not achieve sufficient accuracy in personality prediction. I don't anticipate this to happen because a tiny dataset of handwritten m's was able to have over 80% accuracy determining if someone is worried or not worried.Learn about accountability on Kickstarter
- (7 days)