Article Per Year (5 Year)
p-Index From 2021 - 2026
0.751
P-Index
This Author published in this journals
All Journal Inferensi MATHunesa: Jurnal Ilmiah Matematika
Khusnia Nurul Khikmah
Universitas Palangka Raya
Computer Science & IT Decision Sciences, Operations Research & Management Engineering Mathematics Social Sciences

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
ESG Factors and Interest Rate Stability: A Comparative Analysis of Statistical and Machine Learning Approaches Khusnia Nurul Khikmah; A'yunin Sofro
Inferensi Vol 9 No 1 (2026)
Publisher : Department of Statistics ITS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12962/j27213862.v9i1.9511

Abstract

This study aims to compare the performance of basic statistical approaches and machine learning in identifying environmental, social, and governance (ESG) issues that influence the stability of the rupiah through the Indonesian interest Rate. This study proposes two approaches: machine learning analysis (long short-term memory (LSTM) and multiple long short-term memory (M-LSTM)) and fundamental statistical models (autoregressive integrated moving average (ARIMA), transfer function, and Koyck regression) to forecast the Indonesian interest rate time series data based on ESG factors. The analysis uses a completely randomized design to capture complex patterns, utilizing interest rate data, PM2.5 air quality data, and rainfall index data. This study provides novelty through data splitting scenarios in its empirical analysis and a comparative study of approaches to the model. The analysis results indicate that ESG factors, specifically air quality (PM2.5) and the rainfall index, have a significant influence on Indonesian interest rate values, as determined by the transfer function results, Koyck regression, and M-LSTM model analyses. All proposed model approaches demonstrate good forecasting accuracy, empirically proven by MAPE values with MAPE<10%. The MAPE values of the basic statistical models (Koyck regression, transfer function, and ARIMA) are 0.2842%, 1.0350%, and 2.1245%, outperforming machine learning models (LSTM and M-LSTM) with MAPE values of 4.660% and 7.7353%.
EVALUASI KINERJA PENANGANAN DATA TIDAK SEIMBANG DALAM MEMPREDIKSI LAJU PERTUMBUHAN PENDUDUK DI KALIMANTAN Khusnia Nurul Khikmah; A'yunin Sofro
MATHunesa: Jurnal Ilmiah Matematika Vol. 13 No. 03 (2025)
Publisher : Universitas Negeri Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/mathunesa.v13n3.p332-341

Abstract

Laju pertumbuhan penduduk merupakan indikator demografi krusial yang memengaruhi berbagai aspek kabupaten dan kota di Kalimantan, sehingga membutuhkan analisis komprehensif dalam memodelkan data. Fakta lapangannya masalah ini salah satunya dipengaruhi oleh fenomena ketidakseimbangan data atau satu kategori laju pertumbuhan penduduk lebih dominan. Oleh karena itu, penelitian ini mengusulkan kajian perbandingan metode penanganan data tidak seimbang dengan model analisis regresi logistik. Empat metode penanganan data tidak seimbang ini adalah tanpa penanganan atau baseline, kedua dengan metode random over sampling (ROS), random undersampling (RUS), dan synthetic minority oversampling technique (SMOTE). Dimana data yang digunakan adalah data sekunder yang diambil dari Badan Pusat Statistik (BPS) lima provinsi di Kalimantan. Hasil analisis laju pertumbuhan penduduk kabupaten dan kota di Kalimantan menunjukkan bahwa model regresi logistic biner dengan tanpa penanganan atau baseline memberikan akurasi hasil prediksi terbaik. Berdasarkan nilai akurasi, balanced accuracy, dan ROC menunjukkan nilai tertinggi dibandingkan metode penanganan data tidak seimbang lainnya, yaitu untuk data latih sebesar 66.7%, 53.1%, dan 76.51%. Sedangkan untuk data uji sebesar 72.7%, 62.5%, dan 82.14%.
Performance Evaluation of Clustering Algorithms for Grouping Regencies and Cities in Kalimantan Based on Socio-Demographics Khusnia Nurul Khikmah; A&#039;yunin Sofro
Inferensi Vol 9 No 2 (2026)
Publisher : Department of Statistics ITS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12962/j27213862.v9i2.9512

Abstract

Cluster analysis is an important data exploration technique that aims to identify patterns or characteristics of data, which, in practice, are used to support evidence-based policy making, one of which is used in population and regional development. However, choosing the wrong clustering method can produce useless analysis results. Therefore, this study proposes a comparative study to determine the performance of various clustering algorithms, comparing the performance of hierarchical clustering algorithms (single linkage, complete linkage, average linkage, centroid, ward) and non-hierarchical algorithms (k-means) in regencies and cities in Kalimantan. The study uses secondary data taken from the official websites of the Central Statistics Agency of the five provinces in Kalimantan, where the data to be clustered consists of 56 regencies and cities with eight socio-demographic variables. The analysis results show that all hierarchical methods produce robust and consistent clusters compared to non-hierarchical methods. This decision is based on the silhouette score value 0.701, which is 0.308 better than k-means. This shows that hierarchical methods can clearly identify regencies and cities in Kalimantan based on their socio-demographic factors.
PERBANDINGAN KINERJA MODEL DERET WAKTU ARIMA, LSTM, DAN GRU UNTUK PERAMALAN PM10 HARIAN DI JAKARTA: PERBANDINGAN KINERJA MODEL DERET WAKTU ARIMA, LSTM, DAN GRU UNTUK PERAMALAN PM10 HARIAN DI JAKARTA Khusnia Nurul Khikmah; A'yunin Sofro
MATHunesa: Jurnal Ilmiah Matematika Vol. 14 No. 02 (2026)
Publisher : Universitas Negeri Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/mathunesa.v14n02.p466-472

Abstract

Polusi udara khususnya partikulat matter kategori PM10 merupakan masalah lingkungan utama di kota besar seperti Jakarta. Dampaknya terhadap kesehatan masyarakat dan kebutuhan akan kebijakan mitigasi yang tepat menjadikan analisis kualitas udara sangat mendesak untuk dilakukan. Salah satu pendekatan yang dapat digunakan untuk menganalisis masalah ini adalah melalui peramalan konsentrasi PM10. Penelitian ini bertujuan membandingkan kinerja model statistik tradisional yaitu autoregressive integrated moving average (ARIMA) dengan model deep learning yaitu long short-term memory (LSTM) dan gated recurrent unit (GRU) dalam meramalkan konsentrasi PM10 harian di Jakarta. Data yang digunakan adalah data PM10 periode Januari 2010 hingga Februari 2025 dengan total 5538 observasi harian. Hasil penelitian menunjukkan bahwa model deep learning GRU kemudian LSTM memberikan hasil akurasi yang lebih baik dibanding model statistik tradisional ARIMA, yang dibuktikan secara berurutan melalui nilai RMSE (13,366, 13,441, dan 17,526), MAE (8,918, 8,958, dan 12,416), MSE (178,647, 189,674, dan 307,163), dan MAPE (21,211%, 21,361%, dan 31,228%). Hasil ini menunjukkan bahwa GRU dan LSTM lebih baik dalam menangkap pola non-linear dan ketergantungan jangka panjang pada PM10 dibandingkan metode ARIMA tradisional.
Co-Authors A&#039;yunin Sofro A'yunin Sofro A’yunin Sofro