Prediksi Financial Distress Perusahaan Sektor Non-Finansial di Indonesia Menggunakan Metode Kombinasi Fuzzy C-Means Dan Naïve Bayes Classifier

Balqist, Arsya Auliati (2023) Prediksi Financial Distress Perusahaan Sektor Non-Finansial di Indonesia Menggunakan Metode Kombinasi Fuzzy C-Means Dan Naïve Bayes Classifier. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 10611910000070-Undergraduate_Thesis.pdf] Text
10611910000070-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 April 2025.

Download (2MB) | Request a copy

Abstract

Financial distress adalah keadaan di mana perusahaan mengalami kesulitan keuangan yang dapat menyebabkan kebangkrutan. Jika perusahaan mengalami financial distress dapat merugikan pihak internal maupun eksternal perusahaan. Oleh karena itu, penting untuk memprediksi financial distress pada perusahaan agar tidak menimbulkan kerugian dan juga dapat dijadikan dasar dalam mengambil keputusan atau kebijakan bagi pihak yang membutuhkan. Metode yang digunakan dalam penelitian ini adalah hybrid classifier dan single classifier. Hybrid classifier merupakan metode kombinasi antara Fuzzy C-Means (FCM) dan Naïve Bayes Classifier (NBC), sedangkan single classifier adalah metode Naïve Bayes Classifier. Tujuan yang ingin dicapai dalam penelitian ini adalah untuk mengetahui karakteristik data prediksi financial distress di perusahaan non finansial Indonesia, mengetahui anggota cluster pada data prediksi financial distress di perusahaan non finansial Indonesia, mengetahui tingkat akurasi prediksi financial distress di perusahaan non finansial Indonesia menggunakan metode kombinasi Fuzzy C-Means dan Naïve Bayes Classifier. Hasil dari penelitian menunjukkan bahwa karakteristik data rata-rata rasio keuangan perusahaan yang mengalami financial distress secara garis besar bernilai negatif. Pada tahap clustering menggunakan Fuzzy C-Means terbentuk 2 cluster dengan jumlah anggota cluster 1 adalah kelompok nonfinancial distress dengan 253 anggota perusahaan dan cluster 2 adalah kelompok financial distress dengan 140 anggota. Metode single classifier unggul dalam hal accuracy sebesar 38,42%, specificity sebesar 31,21%, dan AUC sebesar 0,6135. Sedangkan hybrid classifier unggul dalam hal sensitivity sebesar 97,87%. Metode yang terpilih untuk memprediksi perusahaan yang mengalami financial distress adalah metode hybrid classifier karena memiliki hasil sensitivity lebih tinggi dibandingkan dengan metode single classifier
=====================================================================================================================================
Financial distress is a state in which a company experiences financial difficulties that can lead to bankruptcy. If the company experiences financial distress, it can harm both internal and external parties of the company. Therefore, it is important to predict financial distress in the company so as not to cause losses and can also be used as a basis for making decisions or policies for those in need. The method used in this study is a hybrid classifier and single classifier. Hybrid classifier is a combination method between Fuzzy C-Means (FCM) and Naïve Bayes Classifier (NBC). Single classifier is the method of Naïve Bayes Classifier. The purpose of this study is to determine the characteristics of financial distress prediction data in Indonesian non-financial companies, to know the cluster members on financial distress prediction data in Indonesian non-financial companies, to determine the level of accuracy of financial distress prediction in Indonesian non-financial companies using a combination of Fuzzy C-Means and Naïve Bayes Classifier. The results of the study showed that the characteristics of the average data financial ratios of companies experiencing financial distress are broadly negative. At the stage of clustering using Fuzzy C-Means formed 2 clusters with the number of members cluster 1 is a group of nonfinancial distress with 253 members of the company and cluster 2 is a group of financial distress with 140 members. Single classifier method is superior in terms of accuracy of 38.42%, specificity of 31.21%, and AUC of 0.6135. while hybrid classifier excels in terms of sensitivity of 97.87%. The method chosen to predict companies experiencing financial distress is a hybrid classifier method because it has a higher sensitivity than the single classifier method.

Item Type: Thesis (Other)
Uncontrolled Keywords: Financial Distress, Fuzzy C-Means, Naïve Bayes Classifier Financial Distress, Fuzzy C-Means, Naïve Bayes Classifier
Subjects: H Social Sciences > HD Industries. Land use. Labor > HD108 Classification (Theory. Method. Relation to other subjects )
Q Science > Q Science (General) > Q325.5 Machine learning.
Q Science > QA Mathematics > QA278.55 Cluster analysis
Divisions: Faculty of Vocational > 49501-Business Statistics
Depositing User: Arsya Auliati Balqist
Date Deposited: 15 Mar 2023 06:25
Last Modified: 15 Mar 2023 06:25
URI: http://repository.its.ac.id/id/eprint/97766

Actions (login required)

View Item View Item