Comparison Of The Results Of The Jaccard Similarity And KNearest Neighbor Algorithms Using The Case Based Reasoning (CBR) Method On An Expert System For Diagnosing Pediatric Diseases

Altundri Wahyu Hidayatullah, Dian Palupi Rini, Osvari Arsalan

Abstract

Health ranks highest in supporting the continuity of every human activity, especially children. The availability of a doctor is still relatively lacking, especially in remote areas. This makes people have difficulty in diagnosing certain diseases so that medical treatment becomes too late and can even be fatal for the patient. So it is necessary to create a system that has the ability to be able to diagnose diseases in children like an expert. The method used in this study is Case Based Reasoning (CBR) with the Jaccard Similarity Algorithm and K-Nearest Neighbor. Jaccard Similarity is one way to calculate the similarity of two objects (items) which are binary. Similarity calculations are used to generate values whether or not there is a similarity between new cases and existing cases in the case base. While the K-Nearest Neighbor (KNN) Algorithm belongs to the instance-based learning group. The KNN algorithm allows the program to find old cases that are most similar to the current case. Based on the test results using 50 sample data, the expert system can provide diagnostic results in accordance with expert diagnoses. The accuracy results for the K-Nearest Neighbor Algorithm are 72% while the accuracy results for the Jaccard Similarity Algorithm are 70%.

Full Text:

PDF

References

A. F. Indriani, E. Y. Rachmawati, and J. D. Fitriana, “Pemanfaatan Metode Certainty Factor

dalam Sistem Pakar Diagnosa Penyakit pada Anak,” Techno.Com, vol. 17, no. 1, pp. 12–22,

, doi: 10.33633/tc.v17i1.1576.

C. R. Pasalli, V. Poekoel, and X. Najoan, “Sistem Pakar Diagnosa Penyakit Anak

Menggunakan Metode Forward Chaining Berbasis Mobile,” J. Tek. Inform., vol. 8, no. 1,

, doi: 10.35793/jti.7.1.2016.12828.

T. R. Maulidia, “Membuat Sistem Pakar Jauh Lebih Besar Dari Pada Pembuatan Sistem

Biasa . Pakar Digunakan Untuk Memecahkan Masalah Yang Memang Sulit Untuk

Dipecahkan Dengan Pemrograman Biasa , Mengingat Biaya Yang Diperlukan Untuk,”

Coding J. Komput. dan Apl. Untan, vol. 05, no. 03, 2017.

M. Sari, S. Defit, and G. W. Nurcahyo, “Sistem Pakar Deteksi Penyakit pada Anak

Menggunakan Metode Forward Chaining,” J. Sistim Inf. dan Teknol., pp. 130–135, 2020,

doi: 10.37034/jsisfotek.v2i4.34.

A. F. Prayuda, S. Wibisono, and W. Hadikurniawati, “Implementasi Sistem Pakar untuk

Rekomendasi Masakan Tradisional Jawa dengan Metode Case Based Reasoning

Menggunakan Algoritma Similaritas Czekanowski,” Pros. SENDI_U, pp. 978–979, 2018.

F. Fatmayati, - Kusrini, and E. T. Lutfi, “Implementasi Case Base Reasoning Untuk

Mendiagnosa Penyakit Gigi dan Mulut,” Techno.Com, vol. 16, no. 1, pp. 70–79, 2017, doi:

33633/tc.v16i1.1331.

F. D. Wahyudi, D. Remawati, and P. Harsadi, “Sistem Pakar Deteksi Kerusakan Mesin

Bubut Dengan Metode Knn,” J. Teknol. Inf. dan Komun., vol. 6, no. 2, pp. 7–13, 2019, doi:

30646/tikomsin.v6i2.370.

D. A. Pranggono, Sabar, “Sistem Pakar Diagnosa Penyakit Kucing Menggunakan Metode

Forward Chaining (Fc) Berbasis Web,” pp. 1–27, 2017.

A. Gunawan, C. Suhery, and T. Rismawan, “Implementasi Metode Case-Based Reasoning

Dan Similarity Jaccard Coefficient Dalam Identifikasi Kerusakan Laptop,” Coding J.

Komput. dan Apl., vol. 09, no. 02, pp. 292–305, 2021.

A. R. Fabiyanto, Y. T. Mursityo, and D. Pramono, “Pengembangan Sistem Informasi

Penilaian Kinerja Guru Menggunakan Metode Rational Unified Process ( RUP ) Berbasis

Web ( Studi Pada SD Negeri Prigen 1 ),” J. Pengemb. Teknol. Inf. dan Ilmu Komput., vol. 3,

no. 4, pp. 3888–3895, 2019.

DOI: https://doi.org/10.36706/sjia.v5i1.55

Refbacks

  • There are currently no refbacks.