That’s why I didn’t attempt to model gender but rather attraction.
Since gender is more complicated, you would have to reduce the dimension of someone’s “true” gender to this model, losing some information (such as the exact correlation) and projecting them onto multiple points. For example, the lumberjack guy would be projected onto two opposed points in “masculine presenting” and “feminine presenting”. If your attraction area doesn’t encompass both these points you are not attracted to him.
Also, models are simplifications. Since I am not a sociologist, I do not need a highly complex model trying to achieve 99.9% accuracy - nor would I be able to grasp it. A simple one, achieving something like 95% accuracy is sufficient.
That’s why I didn’t attempt to model gender but rather attraction.
Since gender is more complicated, you would have to reduce the dimension of someone’s “true” gender to this model, losing some information (such as the exact correlation) and projecting them onto multiple points. For example, the lumberjack guy would be projected onto two opposed points in “masculine presenting” and “feminine presenting”. If your attraction area doesn’t encompass both these points you are not attracted to him.
Also, models are simplifications. Since I am not a sociologist, I do not need a highly complex model trying to achieve 99.9% accuracy - nor would I be able to grasp it. A simple one, achieving something like 95% accuracy is sufficient.