Article Per Year (5 Year)
p-Index From 2021 - 2026
0.562
P-Index
This Author published in this journals
All Journal Jurnal Gaussian
Deby Fakhriyana
Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro
Other

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
PENERAPAN METODE RANDOM FOREST UNTUK ANALISIS SENTIMEN PENGGUNA APLIKASI BANK DIGITAL SEABANK Fenansia Clara Hana Bangun; Mustafid Mustafid; Deby Fakhriyana
Jurnal Gaussian Vol 15, No 1 (2026): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14710/j.gauss.15.1.36-45

Abstract

Sentiment analysis on reviews of an application becomes an option to see users responses to the service of that particular application. Random Forest is one of the classification modeling techniques that originate from a combination of Decision Trees, providing the final result based on majority voting. This research aims to improve the performance of sentiment classification on customer reviews of Seabank, one of the most widely used digital banking services in Indonesia, by utilizing the Random Forest algorithm. The study involves sentiment analysis of user reviews on the Seabank application, collected from 15,000 reviews on Google Playstore. The review features available on Google Playstore are used as a means to convey opinions as user feedback for an application. Random Forest is trained to classify reviews into 3 sentiment classes: positive, neutral, and negative. Based on the research conducted with model evaluation using Confusion Matrix, an accuracy value of 94.1% was obtained, indicating that Random Forest's accuracy in classifying Seabank customer reviews is 94.1%. This demonstrates the effectiveness of using Random Forest in text review classification due to its high accuracy value.
PENGENDALIAN KUALITAS PUPUK NITROGEN, PHOSPAT, KALIUM (NPK) PELANGI FUSION DI PT PUPUK KALIMANTAN TIMUR MENGGUNAKAN PETA KENDALI MEWMA & MEWMV Bungan Tcania Paulina Grace; Puspita Kartikasari; Deby Fakhriyana
Jurnal Gaussian Vol 15, No 1 (2026): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14710/j.gauss.15.1.110-120

Abstract

The NPK Pelangi Fusion fertilizer is one of the main products manufactured by PT Pupuk Kalimantan Timur. NPK fertilizer is a blend of various plant nutrients, primarily Nitrogen, Phospat, and Kalium, aimed at enhancing crop yields. However, deviations from the NPK fertilizer specifications can lead to inconsistent plant growth and low harvest yields. Statistical Process Control (SPC) is a method used to process data and monitor production processes using statistical techniques, with the goal of detecting changes in process performance through the use of control charts. In this study, the Multivariate Exponentially Weighted Moving Variance (MEWMV) control chart is used to monitor the variance of the production process due to its optimal performance in detecting small variance shifts. Additionally, the Multivariate Exponentially Weighted Moving Average (MEWMA) control chart is used to monitor the mean of the NPK Pelangi Fusion production process, as it quickly detects subtle shifts in variance. The analysis results indicate that the optimal weighting for the MEWMV control chart is ω=0.2 and λ=0.4, resulting in an Average Run Length (ARL) of 370. Similarly, the optimal weighting for the MEWMA control chart is λ=0.06, with an upper control limit of H=9.80 and an ARL of 200. This study concludes that the mean and variance of the multivariate production process have been effectively controlled.
PERAMALAN JUMLAH PENUMPANG PENERBANGAN DOMESTIK PADA TIGA BANDARA UTAMA INDONESIA MENGGUNAKAN MODEL GENERALIZED SPACE TIME AUTOREGRESSIVE (GSTAR) Nabila Zahra Tamami; Puspita Kartikasari; Deby Fakhriyana
Jurnal Gaussian Vol 15, No 1 (2026): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14710/j.gauss.15.1.243-254

Abstract

Forecasting the number of airplane passengers requires a model related to time and location because the data on the number of airplane passengers is space time series data. Generalized Space Time Autoregressive (GSTAR) is an approach to forecast time series data that show a connection between time and location. The GSTAR model can be applied to data with heterogeneous location characteristics.  In this research, the GSTAR model is used to form a forecasting model for the number of domestic flight passengers at three major airports in Indonesia Juanda Airport, Soekarno Hatta Airport, and Hasanuddin Airport, using optimal location weighting. The residual of the GSTAR model satisfies the white noise assumption. The location weights used are uniform weights, inverse distance weights, and cross-correlation normalization. The outcome model is constructed with a first-order differencing, an autoregressive order of 1, and spatial order of 1, resulting in the formation of the GSTAR (11)I(1) model. The most optimal model generated is the GSTAR (11)I(1) model utilizing inverse distance weights due to its smallest sMAPE value compared to other weightings, which stands at 8,71%. This sMAPE value signifies a highly precise forecast accuracy.
Co-Authors Bungan Tcania Paulina Grace Fenansia Clara Hana Bangun Mustafid Mustafid Nabila Zahra Tamami Puspita Kartikasari