![](images/pic01.png)
Linear Regression
For the original Turkish dataset, we chose to do Lasso and Ridge linear regressions to reduce model complexity. Ridge puts constraints on the coefficients which in effect, shrinks the coefficient.
Divorce Prediction Project Made Possible By Machine Learning
Most Americans are well aware that half of marriages end in divorce. The good news is that divorce rate in the US is decreasing, however, so is the marriage rate. According to the Center for Disease Control, the 2018 marriage rate was 6.9 per 1,000 total population while the divorce rate was 2.9 per 1,000 population. Both rates were decreased by a tenth of a point from the previous year and the year before. It seems fewer couples are willing to gamble the risk these days and opt out of marriage. With the high risk of divorce, we found the subject of divorce to be a compelling machine learning exercise. Can divorce be predicted based on specific attributes?
For the original Turkish dataset, we chose to do Lasso and Ridge linear regressions to reduce model complexity. Ridge puts constraints on the coefficients which in effect, shrinks the coefficient.
In addition to linear regression, we also used a heatmap to get an overall look into the distribution of each attribute in our dataset. We used seborn to create the heatmap looking at all 54 survey questions and average values.
For the original Turkish dataset, we chose to do Lasso and Ridge linear regressions to reduce model complexity. Ridge puts constraints on the coefficients which in effect, shrinks the coefficient.
For this next model, a Random Forest was utilized to determine feature importance. This type provides a strong modeling technique that is better than using a single decision tree.
Similar to the Yontem study, our team analyzed the accuracy of three machine learning models on the augmented data set. The models used were K-Nearest Neighbor (KNN), LinearSVC, and SGDClassifier.
Our team used numpy and Plotly to create interactive visuals that reflect the responses of the study at an individual level.
Due to the COVID-19 outbreak, our team of data scientists conducted a pre-recorded final project presentation.
We regret not being able to present in person but have made the link available to you below for your viewing pleasure.