Recently, a student named Bai Li (Li Bai? On Medium, Li Shared how he used logistic regression in ML to help him find a girlfriend. As practical a technique as this, we must observe and learn one.
Let me just say, people have been mistaken about the University of Waterloo in the past. The Waterloo here is in Canada, not the One in Belgium where Napoleon actually took the wheel. The University of Waterloo is a renowned Canadian university and one of the best universities in North America. Its teaching level in mathematics, computer science and engineering is among the top in the world, among which computer Science is ranked 18th in the 2017 USNews World University Rankings.
Good nonsense do not say, we watch the young performance:
The University of Waterloo is known for its lack of social interaction and difficulty in finding a partner. Like me, many young men majoring in computer science thought it was impossible to find a girl, probably until they graduated. Find also don’t know how to find, fall in love again don’t, can only knock on the code to maintain a life like this.
Some people think that love is something that can’t be quantified, that you just have to be yourself. But as a data scientist at the University of Waterloo, I disagree. So I thought, if you’re in computers, why not try machine learning to find a girlfriend?
methodology
Start researching how to use machine learning to find women.
The question at the heart of the study was: What are the attributes that make you stand out from the crowd of men at Waterloo and find a girlfriend?
Many people feel that having money in their pocket makes them more attractive to women, while height and lack of muscle also play a role.
Let’s try to figure out which attributes are the most predictive and which assumptions are not supported by the data.
The following attributes come to mind first:
Dating (target variable) : have a girlfriend, or have been in a relationship for at least six months in the past five years. Nationality: International student (because I am also an international student) Major: CS, SE and ECE Professional Career: Academically successful and found a well-paid internship Interesting: Articulate, always finding interesting things to talk about Social: Outgoing, always wanting to meet new people Confident: Looks confident Height: Taller than me (> 175cm) Glasses: Wear glasses (I wear them too) Fitness: Go to the gym regularly, or exercise Fashion: Pay attention to your appearance and dress tastefully Canada: Have lived in Canada for the last 5 years Asian: From East Asia (because I, too)
As you can see, some of these attributes are very subjective, like what makes a person interesting?
In each of these cases, I assign the values 1 or 0 depending on whether the criteria are met. So, we’re measuring the relationship between these attributes and the objects we can find (in my own sense of the word, of course).
So if you’re looking for an ultra-hardcore, statistical study, this is probably not your cup of tea.
To collect the data, I put everyone I could think of on a table and gave them a score of 0 or 1 for each attribute. Finally, the data set has N=70 rows. If you’ve been at USC for the last two years and know me, chances are you’re on this list.
Analysis of the
First of all, Fisher’s Exact Test is used to analyze the target appointment variable and all explanatory variables, and it is found that three variables have the most significant influence:
Fitness: People who go to the gym or exercise regularly are more than twice as likely to have a girlfriend (P =0.02)
Glasses: People who don’t wear glasses are 70% more likely to have a girlfriend (P =0.08)
Confidence: People who are confident are more likely to have friends (P =0.09)
As I expected, muscular and confident guys were more attractive. But I was surprised by how much of a difference glasses made, and wondered if it was because glasses were associated with being a nerd. So I did some research and found that it was true, there was a research paper that said that most people in both men and women think that wearing glasses makes them less attractive.
Some variables may be predictive of dating success, but it’s hard to pin down because the sample size is small:
Asians have fewer dating opportunities than other ethnic groups. Looking at other factors, although there are fewer women, male computer science majors do not seem to be at a disadvantage
The remaining variables (height/career/fun/social/fashion/place of residence) were not strongly correlated with dating success. I’m sorry, even if you work at Facebook, you don’t have a girlfriend.
Complete results of this experiment:
We then examine the relationship between variables, which helps us identify incorrect model assumptions. Red is positive, blue is negative. We only show correlations with statistical significance <0.1, so most of the relationships between variables are blank.
It looks like there’s a correlation between having a girlfriend, looking confident, going to the gym, not wearing glasses.
Before I go any further, I should emphasize that these friends of mine are not representative of the university of Waterloo as a whole. I met them either in class or at work (all kinds of people, but all computer related) or as acquaintances (from different majors, but mostly from East Asia and living in Canada).
The model trained with these data will also reflect these biases. In the future, I will expand the survey scope and collect more data.
Using logistic regression to predict female ticketing
Wouldn’t it be nice if there was an algorithm that predicted how likely you were to find a woman? Let’s try!
I trained a logistic regression generalized linear model to predict whether there will be women based on the illustrative variables we listed earlier. With the help of the GLmnet and Caret packages in R, I trained this generalized linear model with elastic network regularization. Then, the standard grid search method was used to optimize the hyperparameters, the leave-one cross-validation method was used in each iteration, and the Kappa coefficients were optimized.
The final model’s cross-validation ROC AUC score was 0.673, which means the model is still better at predicting your chances of finding a woman than you would be if you just guessed. Of course, there will always be occasional uncertainties in life, and there will always be surprises.
Anyway, I’m going to the gym and trying to get my glasses off!
Postword: I shared my model in this article to test my chances of finding a mate at the University of Waterloo. However, after opening the link, the model cannot be obtained at present. If you can visit later, we will share this model to predict your “marriage”. Of course, if you can build your own AI that can predict the odds of a romantic love affair, that’s a hoo-hoo!
Learning to find a Girlfriend at the University of Waterloo by Logistic Regression
medium.com/@uw_data_sc…
Follow public accounts
【 Pegasus Club 】
▼
Previous welfare concerns about the pegasus public number, reply to the corresponding keywords package download learning materials; Reply “join the group”, join the Pegasus AI, big data, project manager learning group, and grow together with excellent people!
Microsoft Danniu artificial intelligence series of lessons
(Scan or subscribe)
M.qlchat.com/live/channe… (Qr code automatic recognition)
From beginning to research, the 10 most Readable books in the field of artificial intelligence
RSVP number “2” machine learning & Data Science must-read classic book with resource pack!
Into AI & ML: Learning machine Learning from Basic Statistics (PDF download)
Answer the number “4” to learn about ARTIFICIAL intelligence, 30 books should not be missed (with electronic PDF download)
Answer number “6” AI AI: 54 Industry Blockbuster Reports
TensorFlow Introduction, Installation tutorial, Image Recognition application (with installation package/guide)
Reply number “12”
Small white | Python + + machine learning Matlab neural network theory + practice + + + depth video + courseware + source code, download attached!
Reply number “14” small white | machine learning and deep learning required books + machine learning field video/PPT + large data analysis books recommend!
17 mind maps for machine learning statistics
Ten years ago on This day on Machine Learning Projects.
Machine learning: How to go from beginner to Never Giving up? (With benefits)
Respond to digital “24” flash download | 132 g programming data: Python, JAVA, C, C + +, robot programming, PLC, entry to the proficient in ~
Reply number “25” limited resources | 177 g Python/machine learning/TensorFlow video/deep learning algorithm, introduction to cover/intermediate/project each stage!
Reply number “26” introduction to artificial intelligence book list recommended, learn AI please collect well (attached PDF download)
Reply | digital “27” Wu En of Stanford CS230 deep learning course a full range of information release (download)
Reply number “28” Programmers who understand this technology are being snapped up by BAT… (Information pack included)
Respond to digital “29” dry | 28 this big data/data analysis, data mining ebook collection of free download!
Reply digital “30” receive | 100 + artificial intelligence study, deep learning, machine learning, big data, algorithms such as data, decisive collection!
FMI Artificial Intelligence and Big Data Summit Guest Speech PPT
Top 10 AI Jianghu Fields
Machine Learning Practical Experience Guide
More than 100 Papers on deep Learning
Top ten Classic Algorithms of Data Mining
6.10 Ele. me & Pegasus Project Management Practice PPT