Many years ago I wrote a popular  (for an academic)  article, “Measuring and Managing Technological Knowledge.” The basic idea is that some  concepts are well understood, many others are not, and over time the tendency is to move from poorly understood (crafts) to well understood (science). Anyway, in class I used the example of romance to prove that this model is very general. “When you were 14, you had absolutely no idea how to impress a girl. When you were 20, you at least knew what the key variables are, even though you didn’t know how to make them happen reliably.” Etc. (Another example is the increasingly scientific business of prostitution – but I won’t tell that one here, and I doubt I had courage to tell it in class.)
Anyway, technological knowledge continues to progress – in this case, the technology of finding romantic partners. (The earliest technology for this was matchmakers e.g. in India, who are still, I suspect, quite good in some situations.) First-generation e-dating, according to this article, was arguably worse than human matchmaking – it tried to use a mathematical approach, which was based on completely inadequate scientific understanding of attraction. But there are several measurement methods on the horizon that will change this. (Know how  to measure is stage 3 of my 7 stages of knowledge.)

Here is an article describing some of this research.

Let’s get the basics over with,” W said to M when they met on a 4-minute speed date. “What are you studying?” “Uh, I’m…

Source: Finding Your Soul Mate Online May Be Harder Than the Dating Sites Suggest

P.S. Talking about matchmakers reminds me of a friend of a friend who served as matchmaker for a major social event at Wellesley. She did a pretty good job, but I was too ignorant to follow up with the girl after the dance, and ….