A hybrid model to study the demographic profile of women in view of assisted reproductive techniques through machine learning models

Assisted Reproductive Techniques (ART) refers to the reproductive measures that address the issues like infertility, low sperm count, unable to conceive and help the couples to achieve pregnancy. Various Assisted Reproductive techniques like in-vitro Fertilization, Intrauterine Insemination are used to treat patients with infertility. This paper studies the demographic profile of women in view of Assisted Reproductive Techniques through Machine Learning Models. Artificial Intelligence is a vast field that has proved its worth in almost all areas. Machine learning is a branch of AI where computers learn from the data provided and improve their performance by learning from the data provided. Demographic profile of Women is important to study from the view of ART methods as this would make machine understand whether the women is healthy or not. She needs to change her lifestyle or opt for Assisted Reproductive Techniques to achieve a successful conception. In this paper, Machine Learning models are applied on Primary data i.e. Data is collected from women of age 25–60 and profile of women is studied to achieve the healthy and non-healthy status of women. The study evaluates Logistic Regression, Decision Tree and Linear Discriminant Analysis, demonstrating that the hybrid approach achieves the highest accuracy (for Healthy Women: F1 score is: 0.80, Precision is: 1.00 and recall is 0.67), (for women with health issue F1 score is 0.93, Precision is 0.88 and Recall is 1.00). Further the study highlights the importance of women health and important factors that decide whether women need Assisted Reproductive Methods to achieve conception or not.