Divorce Prediction with Machine Learning - Insights and LIME Interpretability

Aug 5, 2026

Divorce is a prevalent social issue in developed countries, with nearly half of recent marriages ending in involuntary divorce or separation and causing serious consequences for mental health and personal life. This work predicts divorce using a divorce predictor dataset and six machine learning algorithms: Logistic Regression, Linear Discriminant Analysis, K-Nearest Neighbors, CART, Gaussian Naive Bayes, and Support Vector Machines. Results indicate SVM, KNN, and LDA achieve up to 98.57% accuracy, while LIME provides detailed interpretability of prediction probabilities. A divorce predictor app is built using ten key features.

LanguageEnglish

CategoryTechnology

FormatPDF

Pages5

Reading Time13 min

AccessFree to download

Introduction

Prior research on divorce prediction

Dataset and prediction task

Machine learning methods

Model evaluation results

LIME interpretability

Local explanations and probability analysis

Divorce predictor application

Ten key features

Divorce Prediction with Machine Learning: Insights and LIME Interpretability Md Manjurul Ahsan School of Industrial and Systems Engineering University of Oklahoma Norman, OK 73019, USA [ahsan@ou.edu](ahsan@ou.edu) arXiv :2310 .08620v1 [ cs .LG] 12 Oct 2023 Abstract—Divorce is one of the most common social issues in developed countries like in the United States. Almost 50% of the recent marriages turn into an involuntary divorce or separation. While it is evident that people vary to a different extent, and even over time, an incident like Divorce does not interrupt the individual’s daily activities; still, Divorce has a severe effect on the individual’s mental health, and personal life. Within the scope of this research, the divorce prediction was carried out by evaluating a dataset named by the ’divorce predictor dataset’to correctly classify between married and Divorce people using six different machine learning algorithms- Logistic Regression (LR), Linear Discriminant Analysis (LDA), K-Nearest Neighbors (KNN), Classification and Regression Trees (CART), Gaussian Nave Bayes (NB), and, Support Vector Machines (SVM). Preliminary computational results show that algorithms such as SVM, KNN, and LDA, can perform that task with an accuracy of 98.57% . This work’s additional novel contribution is the detailed and comprehensive explanation of prediction probabilities using Local Interpretable Model-Agnostic Explanations (LIME). Utilizing LIME to analyze test results illustrates the possibility of differentiating between divorced and married couples. Finally, we have developed a divorce predictor app considering ten most important features that potentially affect couples in making decisions in their divorce, such tools can be used by any one in order to identify their relationship condition. Index Terms—Machine Learning, Statistical Analysis, SVM, KNN I. INTRODUCTION Divorce also is known as the dissolution of marriage, which is the process of terminating a marriage or marital union [1] . While it seems as simple as a separation between two people, a divorce may have a significant effect on an individual. A study conducted by the Institute for Population Research [2] showed that around 40% of American children usually experience parental divorce/separation during their childhood and has a long-term effect on their mental health. Parallel research identifies that children from divorced families received less financial support compared to other children during their lifetime. Additionally, those children also felt more distress due to a lack of emotional support from the families [3] . As a result, parental divorce influences child behavior ina negative manner, which leads to anger, frustration, and depression. A divorce could significantly hamper couples, personal and social life. Duncan & Hoffman (1985) and Morgan (1989) [4], [5], showed that individuals’ separation also confronted a variety of stress, financial crisis, reduction in social networks and moving on (Amato, 2000;McLanahan & Sandefur, 1994) [6], [7] . Augustine (2000) found that men were nearly 4.8 times as likely to commit suicide as women. They have used the NLMS data for their study. Their study demonstrated a potential relationship between marital status with suicidal risk and,according to the Centers for Disease Control and Prevention (2017), On average, 1.4 million American attempted suicide. The combined medical and work loss are observed as 69 billion dollars [2] . Apart from this, the relationship between divorce and the individuals driving performance such as accident rate, speeding, failure to yield etc. are also highly correlated [8] . In a nutshell, divorce has a significant effect, which brings damage to various aspects such as social, economic, physical, mental and so on. During the past few decades, the effect of divorce and the reason behind the divorce/marriage, have been studied thoroughly, while very few studies focused on foreca

FAQs

Which algorithms are used for divorce prediction in the study?

The study evaluates Logistic Regression, Linear Discriminant Analysis, K-Nearest Neighbors, Classification and Regression Trees, Gaussian Naive Bayes, and Support Vector Machines.

What accuracy do the best-performing models achieve?

Preliminary results show SVM, KNN, and LDA can reach an accuracy of 98.57% for the classification task.

How does LIME contribute to the research?

LIME is used to explain prediction probabilities by generating local, interpretable explanations, helping distinguish between divorced and married couples.

Aug 5, 2026

Upload by Ava Thompson