Speed Dating @ Machine Learning

Table of contents

• Introduction
• About the Data
• Project Overview
• Setup
• Result

---

1. Introduction

In today’s busy world, finding and dating a romantic partner seems more time consuming than ever. As a result, many people have turned to speed dating as a solution that allows one to meet and interact with a large number of potential partners in a short amount of time. In this report, we want to explore what people are looking for in their speed dating matches, what it takes to become successful in getting approvals from a potential partner, if there exist any gender differences, and if any other factors (such as the order you met your partner) influence peoples’ decisions. Finally, we’d like to determine if people really know what they want by comparing their self-reported answers to what actually influences peoples’ decisions.

---

2. About the Data

The data set we will explore in the project is named Speed Dating Experiment, as found on Kaggle.com. It was compiled by professors Ray Fisman and Sheena Iyengar from Columbia Business School, originally used for their paper Gender Differences in Mate Selection: Evidence From a Speed Dating Experiment. And later for Racial Preferences in Dating It was generated from a series of experimental speed dating events from 2002 to 2004 and includes data related to demographics, dating habits, lifestyle information, an attribute evaluation questionnaire taken when the participants sign up, and each participant’s ratings for others during the 4 minute interactions. Finally, individuals were asked if they would like a second date with their partners and rated again on similar questions after the event, when matches have met with each other and dated for several times.

---

3. Project Overview

The aim of this project is to get a hands-on practice on typical workflows usually seen in any machine learning project. Below mentioned is the basic workflow for this project:

• Exploratory Data Analysis (EDA)
• Pre-processing
• Modelling
• Test Analysis

Note: For more details, please refer to the notebook for each part. For visualizations, you can have a look at this

---

4. Setup

We have a pretty straightforward setup.

1. Set up a new Python env

If you have conda installed, run the following command

conda create -n machine_learning python=3.8 -y

2. Now clone the repository

git clone https://github.com/mohammadzainabbas/Speed-Dating-ML.git
cd Speed-Dating-ML/

3. Install the required dependencies

conda activate machine_learning
pip install -r requirements.txt

Now, everything that you need is installed and ready to go.

---

5. Result

Model	Accuracy	Precision	Recall	F1 score (macro)
Logistic Regression	0.754	0.721	0.678	0.745
Support Vector Machine	0.747	0.713	0.674	0.738
k-Nearest Neighbours	0.665	0.630	0.511	0.644
Gradient Boosting	0.755	0.725	0.680	0.746
Voting Ensemble	0.751	0.739	0.634	0.738
Stacking Ensemble	0.756	0.726	0.679	0.747

---