Perbandingan kinerja algoritma Hopfield, Random Forest dan Support Vector Machine untuk deteksi dini penyakit diabetes

Dewi, Nurul Aulia (2023) Perbandingan kinerja algoritma Hopfield, Random Forest dan Support Vector Machine untuk deteksi dini penyakit diabetes. Sarjana thesis, UIN Sunan Gunung Djati Bandung.

[img]
Preview
Text (COVER)
1_cover.pdf

Download (40kB) | Preview
[img]
Preview
Text (ABSTRAK)
2_abstrak.pdf

Download (89kB) | Preview
[img]
Preview
Text (DAFTAR ISI)
3_daftarisi.pdf

Download (92kB) | Preview
[img]
Preview
Text (BAB I)
4_bab1.pdf

Download (90kB) | Preview
[img] Text (BAB II)
5_bab2.pdf
Restricted to Registered users only

Download (515kB) | Request a copy
[img] Text (BAB III)
6_bab3.pdf
Restricted to Registered users only

Download (367kB) | Request a copy
[img] Text (BAB IV)
7_bab4.pdf
Restricted to Registered users only

Download (374kB) | Request a copy
[img] Text (BAB V)
8_bab5.pdf
Restricted to Registered users only

Download (21kB) | Request a copy
[img] Text (DAFTAR PUSTAKA)
9_daftarpustaka.pdf
Restricted to Registered users only

Download (153kB) | Request a copy

Abstract

INDONESIA: Perbandingan kinerja algoritma merupakan suatu proses yang bertujuan untuk mengevaluasi serta membandingkan efektivitas berbagai algoritma dalam menjalankan tugas klasifikasi pada data yang ada. Tujuan utamanya yaitu mengidentifikasi algoritma yang paling sesuai atau memberikan hasil terbaik dalam mengklasifikasikan data berdasarkan fitur atau atribut yang dimiliki. Proses ini menjadi penting karena dalam banyak kasus, penelitian cenderung menggunakan algoritma yang umum daripada mengeksplorasi potensi dari algoritma-algoritma lainnya. Supaya penelitian ini lebih terarah, dilakukan pendekatan dengan membandingkan tiga algoritma: Hopfield, Random Forest, dan Support Vector Machine (SVM), dalam konteks deteksi dini penyakit diabetes. Pendekatan ini memanfaatkan metode Knowledge Discovery in Databases (KDD) untuk mengembangkan sistem yang bertujuan untuk menggali potensi dari data yang diambil dari dataset. Penelitian ini melibatkan perbandingan antara data latih dan data uji dengan perbandingan 5 variasi: 50:50, 60:40, 70:30, 80:20, dan 90:10. Hasil penelitian menunjukkan bahwa algoritma Random Forest mencapai akurasi tertinggi pada perbandingan data 90:10, dengan nilai sebesar 88,1%. Selain itu, algoritma SVM juga menghasilkan kinerja baik dengan akurasi 87,6% pada variasi data 50:50. Sementara itu, algoritma Hopfield mencapai akurasi sebesar 85% pada variasi data 70:30. ENGLISH: The comparison of algorithm performance is a systematic procedure designed to assess and contrast the efficacy of diverse algorithms in executing classification tasks on pre-existing datasets. The primary objective is to ascertain which algorithm is the most suitable or yields the most favorable outcomes in classifying data based on its inherent features or attributes. This process holds significant importance as, in numerous instances, research endeavors tend to rely on conventional algorithms, often neglecting the potential of alternative ones. To streamline the research focus, an approach was adopted that involved the comparative analysis of three distinct algorithms: Hopfield, Random Forest, and Support Vector Machine (SVM), within the context of early diabetes detection. This approach leveraged the Knowledge Discovery in Databases (KDD) methodology to construct a system with the intent of delving into the latent potential of the data extracted from the dataset. The research encompassed the evaluation of training and test data across five distinct ratios: 50:50, 60:40, 70:30, 80:20, and 90:10. The findings revealed that the Random Forest algorithm exhibited the highest accuracy, achieving an 88.1% accuracy rate in the 90:10 data ratio. Furthermore, the SVM algorithm demonstrated commendable performance, attaining an 87.6% accuracy rate in the 50:50 data variation. Meanwhile, the Hopfield algorithm achieved an accuracy rate of 85% in the 70:30 data variation.

Item Type: Thesis (Sarjana)
Uncontrolled Keywords: Perbandingan Algoritma; Algoritma Hopfield; Random Forest; Support Vector Machine (SVM); Knowledge Discovery Database; Klasifikasi
Subjects: Special Computer Methods
Engineering
Divisions: Fakultas Sains dan Teknologi > Program Studi Teknik Informatika
Depositing User: Nurul Aulia Dewi
Date Deposited: 12 Sep 2023 02:12
Last Modified: 12 Sep 2023 02:12
URI: https://digilib.uinsgd.ac.id/id/eprint/76681

Actions (login required)

View Item View Item