University of Kashan Soft Computing Journal 2322-3707 10 2 2022 02 20 Prediction of couple relationship during the Covid-19 period using correlation-based feature selection and machine learning Prediction of couple relationship during the Covid-19 period using correlation-based feature selection and machine learning 56 71 112028 10.22052/scj.2022.243472.1040 FA Ali Asghar Akhavan Mahdavi Department of Computer Engineering, Khavaran Institute of Higher Education, Mashhad, Iran. Elham Mahdipour Department of Computer Engineering, Khavaran Institute of Higher Education, Mashhad, Iran. Journal Article 2022 01 25 During the COVID-19 pandemic, many societies faced an increase in divorce rates due to dissatisfaction, incompatibility, and turmoil in couples' relationships. One of the effective factors in a successful marriage is realistic expectations, and examining factors that contribute to successful marriages is an important step towards achieving this goal. In this regard, this paper aims to investigate the effective factors in successful marriages or divorce of couples using machine learning algorithms and feature selection based on discovering correlations between features. A standardized questionnaire with 54 questions is developed based on the Gottman couple therapy questionnaire available on Kaggle. The goal is to predict the type of couple relationship and detect the survival of the relationship using the analysis of 170 couples' responses. The questionnaire is designed in Persian and distributed among 33 Iranian families. Machine learning algorithms are used to predict the type of relationship between couples. Moreover, the effective features of couples’ relationships in Iranian culture are extracted using Pearson's correlation coefficient and compared with effective features in another study conducted on individuals from different cultures. The results showed that cultural differences can change some of the effective features. The experimental results indicate that the most effective factors in predicting whether a person is divorced or married are (1) intimacy between couples and (2) silence and suppression of anger during discussions and disputes. Among the examined machine learning algorithms, the decision tree and naive Bayes algorithms have the highest efficiency, with 100% accuracy. During the COVID-19 pandemic, many societies faced an increase in divorce rates due to dissatisfaction, incompatibility, and turmoil in couples' relationships. One of the effective factors in a successful marriage is realistic expectations, and examining factors that contribute to successful marriages is an important step towards achieving this goal. In this regard, this paper aims to investigate the effective factors in successful marriages or divorce of couples using machine learning algorithms and feature selection based on discovering correlations between features. A standardized questionnaire with 54 questions is developed based on the Gottman couple therapy questionnaire available on Kaggle. The goal is to predict the type of couple relationship and detect the survival of the relationship using the analysis of 170 couples' responses. The questionnaire is designed in Persian and distributed among 33 Iranian families. Machine learning algorithms are used to predict the type of relationship between couples. Moreover, the effective features of couples’ relationships in Iranian culture are extracted using Pearson's correlation coefficient and compared with effective features in another study conducted on individuals from different cultures. The results showed that cultural differences can change some of the effective features. The experimental results indicate that the most effective factors in predicting whether a person is divorced or married are (1) intimacy between couples and (2) silence and suppression of anger during discussions and disputes. Among the examined machine learning algorithms, the decision tree and naive Bayes algorithms have the highest efficiency, with 100% accuracy. https://scj.kashanu.ac.ir/article_112028_89bc23205d09cb883ae673c993bc895f.pdf