Adoption of Artificial Intelligence and Machine Learning Algorithms on Assessment and Prevalence of Cervical Cancer Risk Factor Amongst Women in Umuagwo Ohaji/Egbema LGA Imo State, Nigeria

• Author(s): Oguoma Ikechukwu Stanley ; Ebele Precious Okemba ; Shanice Kristy Archibald ; David Osondu Akuchie ; Ajero Chukwuka Evans
• Paper ID: 1707947
• Page: 919-932
• Published Date: 25-04-2025
• Published In: Iconic Research And Engineering Journals
• Publisher: IRE Journals
• e-ISSN: 2456-8880
• Volume/Issue: Volume 8 Issue 10 April-2025

Abstract

This study was able to ascertain clearly how cervical cancer could be managed in Umuagwo, Ohaji/Egbema L.G.A Imo State using a hybrid predictive model with outstanding focus on three different variables for each algorithm applied on the dataset which are (DT= motivation_willingness, Perception severity, empowerment knowledge and RF= empowerment ability, perception severity, motivation strength). The major objectives of the research includes to analyze the dataset using two major classification algorithms tool to predict or identify the risk factor of cervical cancer amongst women of Umuagwo Ohaji/Egbema LGA and to build and train a hybrid model that can analyze and predict the major cause of cervical cancer amongst women of childbearing age in Umuagwo Ohaji/Egbema LGA. The research was motivated in other to identify the strong believe the people of Umuagwo in Ohaji/Egbema has on their deity/juju as being responsible for the frequent death of their women. The study employed two machine learning classification algorithm methods which include: Decision Tree (DT) and Random Forest (RF) algorithms. The data was analyzed with R and JASP platform while the experiments are done on the dataset containing TRAIN, VALIDATION, TEST) for Random Forest (RF) with a percentage rate of 46%, 12% and 14% respectively and for Decision tree (DT) with percentage rate of (TRAIN 58% and TEST 14%) making up to the total of 72% for both algorithms. The results prediction accuracy was concluded by comparing the two developed models involving their different F1 scores, confusion matrix, Evaluation Metrics, Roc Curves, Precision (positive predictive value)/support and the RF out-of-bag result shown in Table 9 of this paper. From the predicted result shown by the hybrid model, it was observed that lack of willingness or motivation, empowerment ability and knowledge was the major risk factor causing the increase of cervical cancer amongst women of Umuagwo Ohaji/Egbema whereby leading them to frequent death and not their deity/juju as perceived by the villagers with percentage accuracy of DT= ( 0(NO) = 0.824 : 82% Accuracy and 1(YES) = 0.727 :73% ) while RF= ( 0(NO) = 0.889: 89% Accuracy and 1(YES) = 0.800: 80% Accuracy) respectively.

Keywords

Artificial Intelligence, Machine Learning, Decision Tree and Random Forest Classification Models, Cervical Cancer Risk Factor amongst Women, health diagnosis and treatment and cervical cancer Prevalence.

Citations

IRE Journals:
Oguoma Ikechukwu Stanley , Ebele Precious Okemba , Shanice Kristy Archibald , David Osondu Akuchie , Ajero Chukwuka Evans "Adoption of Artificial Intelligence and Machine Learning Algorithms on Assessment and Prevalence of Cervical Cancer Risk Factor Amongst Women in Umuagwo Ohaji/Egbema LGA Imo State, Nigeria" Iconic Research And Engineering Journals Volume 8 Issue 10 2025 Page 919-932

IEEE:
Oguoma Ikechukwu Stanley , Ebele Precious Okemba , Shanice Kristy Archibald , David Osondu Akuchie , Ajero Chukwuka Evans "Adoption of Artificial Intelligence and Machine Learning Algorithms on Assessment and Prevalence of Cervical Cancer Risk Factor Amongst Women in Umuagwo Ohaji/Egbema LGA Imo State, Nigeria" Iconic Research And Engineering Journals, 8(10)