Rancang Bangun Chatbot Monitoring Menstruasi Berbasis Decision Tree C4.5 untuk Deteksi Dini Polycystic Ovarian Syndrome (PCOS)

Mediana, Trisha (2023) Rancang Bangun Chatbot Monitoring Menstruasi Berbasis Decision Tree C4.5 untuk Deteksi Dini Polycystic Ovarian Syndrome (PCOS). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 07311940000049-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (3MB) | Request a copy

Abstract

Hasil analisis terhadap 1,6 juta data riwayat siklus menstruasi menunjukkan wanita usia 18-24 tahun memiliki angka ketidakteraturan siklus menstruasi dua kali lebih besar dari wanita usia 35-39 tahun. Selain itu, pengetahuan mengenai keseahtan reproduksi masih belum merata pada perempuan usia produktif. Sedangkan menstruasi tidak teratur merupakan salah satu gejala utama polycystic ovarian syndrome (PCOS). Sebelum diagnosis definitif PCOS, remaja wanita dengan tanda-tanda klinis kelebihan androgen dan oligomenorea/amenore, dianggap "berisiko" menderita PCOS. Wanita penderita PCOS memiliki peningkatan risiko intoleransi glukosa, diabetes mellitus tipe 2, hepatic steatosis, sindrom metabolisme, hipertensi, dislipidemia, penggumpalan pada vascular, insiden cerebrovascular, cardiovascular event, subfertilitas, komplikasi obstetri, endometrial atypia atau carcinoma, kanker ovarium, kelainan mood dan psikoseksual. Diusulkan suatu sistem chatbot berbasis self assesment dengan memanfaatkan arsitektur microservice dan fungsi monitoring dan pengingat siklus menstruasi serta deteksi dini risiko PCOS menggunakan penerapan Decision Tree C4.5 yang disertai website edukasi seputar menstruasi dan PCOS untuk mengurangi tingkat keterlambatan diagnosis. Sistem diujikan terhadap 46 pengguna wanita pada sistem ini terdiri terdiri atas fitur monitoring dengan memperhitungkan perkiraan tanggal menstruasi yang akan datang, fitur pengingat hari menstruasi saat mendekati atau hari-H menstruasi, fitur self-assessment untuk deteksi risiko PCOS menggunakan decision Tree C4.5 dengan kurang lebih hasil rata-rata akurasi diagnosis risiko PCOS oleh decision tree C4.5 sebesar 94,59% yang lebih baik dari decision tree ID3 rata-rata akurasi 91,65%. Penerapan modifikasi pruning pada decision tree C4.5 mempersingkat proses klasifikasi kemudian penilaian performa sistem secara keseluruhan menunjukkan rata-rata performa sudah baik dan berguna menurut survei dari 35 orang pengguna
=================================================================================================================================
The results of an analysis of 1.6 million menstrual cycle history data show that women aged 18-24 years have a menstrual cycle irregularity rate twice as large as women aged 35-39 years. In addition, knowledge about reproductive health is still uneven among women of productive age. Meanwhile, irregular menstruation is one of the main symptoms of polycystic ovary syndrome (PCOS). Prior to the definitive diagnosis of PCOS, adolescent women with clinical signs of androgen excess and oligomenorrhea/amenorrhea were considered at risk of PCOS. Women with PCOS have an increased risk of glucose intolerance, type 2 diabetes mellitus, hepatic steatosis, metabolic syndrome, hypertension, dyslipidemia, vascular clots, cerebrovascular incidents, cardiovascular events, subfertility, obstetric complications, endometrial atypia or carcinoma, ovarian cancer, mood and psychosexual disorders. A self-assessment-based chatbot system is proposed by utilizing microservice architecture and the function of monitoring and reminding the menstrual cycle and early detection of PCOS risk using the application of Decision Tree C4.5 accompanied by an educational website about menstruation and PCOS to reduce the delayed diagnosis rate of PCOS . The system was tested on 46 female users in this system consists of a monitoring feature by calculated the estimation date of the upcoming menstruation, a reminder feature of the menstrual day when approaching or the D-day of menstruation, a self-assessment feature for PCOS risk detection used Decision Tree C4.5 with approximately the average result of PCOS risk diagnosis accuracy by decision tree C4.5 of 94.59% which is better than the ID3 decision tree with an average accuracy of 91.65%. The application of pruning modifications to the C4.5 decision tree shortens the classification process then the overall system performance assessment showed the average performance was good and useful according to a survey of 35 users

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Chatbot, Menstruation, PCOS, Menstruasi
Subjects:	R Medicine > RG Gynecology and obstetrics
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User:	Trisha Mediana
Date Deposited:	01 Dec 2023 02:26
Last Modified:	01 Dec 2023 02:26
URI:	http://repository.its.ac.id/id/eprint/103670

Actions (login required)

View Item