Article Per Year (5 Year)
p-Index From 2020 - 2025
1.066
P-Index
This Author published in this journals
All Journal Jurnal Statistika Universitas Muhammadiyah Semarang Jurnal Teknologi Informasi dan Ilmu Komputer BAREKENG: Jurnal Ilmu Matematika dan Terapan Southeast Asian Journal of Islamic Education SEMINAR NASIONAL PENELITIAN DAN PENGABDIAN KEPADA MASYARAKAT
Lusia, Dwi Ayu
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Energy Engineering Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Mathematics Mechanical Engineering Nursing Physics Transportation Other

Published : 6 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 6 Documents
Search

 1 
Perbandingan Metode Jaringan Saraf Tiruan, Fuzzy, Dan Anfis Pada Peramalan Data Inflasi Indonesia Lusia, Dwi Ayu; Semathea, Karen; Sumarminingsih, Eni; Efendi, Achmad
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 12 No 3: Juni 2025
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.2025128613

Abstract

Peramalan adalah teknik penting untuk mengestimasi nilai masa depan berdasarkan data historis. Namun, metode peramalan sering menghadapi tantangan dalam memilih model dengan tingkat akurasi terbaik. Penelitian ini bertujuan membandingkan kinerja metode Jaringan Syaraf Tiruan (JST) dan Fuzzy Metode Sugeno serta gabungan kedua metode yang disebut Adaptive Neuro Fuzzy Inference System (ANFIS). Ketiga metode digunakan untuk meramalkan inflasi bulanan Indonesia. Penerapan ketiga metode membutuhkan penentuan input yang berdasarkan stasioner dan PACF. Data tidak stasioner lag 2 sehingga Differencing lag 2 kemudian tidak ada lag yang keluar pada PACF. Berdasarkan kedua hal tersebut ditentukan inputnya ialah  dan . Hasil menunjukkan bahwa metode JST dengan 3 lapisan tersembunyi dengan banyak neuron (2,1,1) memberikan kinerja terbaik (nilai RMSE terkecil sebesar 1,16127 pada data testing). Metode terbaik tersebut digunakan untuk meramalkan Inflasi bulan September 2023 hingga Desember 2024 cenderung konstan antara 2,68879% hingga 2,68887%. Kontribusi riset ini adalah metode advance (ANFIS) dengan menggabungankan dua metode (JST dan Fuzzy) belum tentu lebih baik daripada metode tanpa penggabungan (JST atau Fuzzy).   Abstract Forecasting is an important technique for estimating future values ​​based on historical data. However, forecasting methods often face challenges in choosing a model with the best level of accuracy. This study aims to compare the performance of the Artificial Neural Network (ANN) and Fuzzy Sugeno Method methods and a combination of the two methods called the Adaptive Neuro Fuzzy Inference System (ANFIS). The third method is used to predict Indonesia's monthly inflation. The application of the third method requires input determination based on stationary and PACF. The data is not stationary lag 2 so that Differencing lag 2 then there is no lag that comes out in PACF. Based on these two things, the input is determined to be Y_(t-1) and Y_(t-2). The results show that the ANN method with 3 hidden layers with many neurons (2,1,1) gives the best performance (the smallest RMSE value is 1.16127 on the test data). The best method used to predict inflation from September 2023 to December 2024 tends to be constant between 2.68879% to 2.68887%. The contribution of this research is that the advanced method (ANFIS) by combining two methods (ANN and Fuzzy) is not necessarily better than the method without combining (ANN or Fuzzy).
CLUSTERING DISTRICTS/CITIES IN EAST JAVA PROVINCE BASED ON HIV CASES USING K-MEANS, AGNES, AND ENSEMBLE Lusia, Dwi Ayu; Salsabila, Imelda; Kusdarwati, Heni; Astutik, Suci
BAREKENG: Jurnal Ilmu Matematika dan Terapan Vol 19 No 1 (2025): BAREKENG: Journal of Mathematics and Its Application
Publisher : PATTIMURA UNIVERSITY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30598/barekengvol19iss1pp63-72

Abstract

Cluster analysis is a method of grouping data into certain groups based on similar characteristics. This research aims to group districts/cities in East Java Province in 2021 based on HIV cases using hierarchical cluster analysis (AGNES), non-hierarchical cluster analysis (K-means), and ensemble clustering. The study found that the ensemble clustering solution forms four clusters, consistent with the results of AGNES clustering. This suggests that ensemble clustering improves the quality of cluster solutions by leveraging both hierarchical and non-hierarchical methods. The grouping of districts/cities based on HIV cases provides a clear distribution pattern for more targeted interventions. The study is limited to HIV cases in East Java Province and may not be generalizable to other regions with different epidemic characteristics. Additionally, the study focuses on clustering methods without investigating temporal changes in HIV case distribution. This research is one of the few studies that applies ensemble clustering to HIV cases in East Java Province. It combines hierarchical and non-hierarchical methods to improve the clustering process and provides a practical approach for regional HIV control planning.
COMPARISON OF FEEDFORWARD NEURAL NETWORK AND LONG SHORT TERM MEMORY IN SENTIMENT ANALYSIS OF SHOPEE APPLICATION REVIEWS Lusia, Dwi Ayu; Simanjuntak, Yessica Maretha
Jurnal Statistika Universitas Muhammadiyah Semarang Vol 13, No 1 (2025): Jurnal Statistika Universitas Muhammadiyah Semarang
Publisher : Department Statistics, Faculty Mathematics and Natural Science, UNIMUS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26714/jsunimus.13.1.2025.74-84

Abstract

Sentiment analysis is a method for generating types of views or opinions that express positive, neutral or negative sentiments. The application of sentiment analysis was carried out to determine the sentiment of Shopee application users. This research uses an artificial neural network algorithm to learn patterns from training data to predict the sentiment of the test data class. The aim of the research is to determine sentiment classification, identify the optimal Feedforward Neural Network and Long Short Term Memory architectural models in classifying user reviews of the Shopee application and compare the performance of the models based on the level of accuracy. The data set is divided into training data and test data respectively by 80% and 20%. The research results showed that there were 91.865 reviews with positive sentiment, 63.038 negative reviews and 26.662 neutral reviews based on Valanced Aware Dictionary Sentiment lexicon dictionary. The network architecture used is one hidden layer, with 137 hidden neurons and a two hidden layer model, with 491 units of first hidden neurons and 38 units of second hidden layer neurons. Evaluation of sentiment classification of Shopee application users resulted in the highest accuracy rate on the single-layer LSTM model, at 68,93%, with precision of 61,29%, and recall of 56,10%.
Desain Faktorial untuk Pembuktian Teori Masters dalam Penentuan Jumlah Lapisan dan Neuron Tersembunyi pada Peramalan Multivariat dengan Jaringan Syaraf Tiruan Lusia, Dwi Ayu; Ambarwati, Awalludiyah
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 7 No 1: Februari 2020
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Jaringan syaraf tiruan merupakan metode yang sangat sering digunakan untuk peramalan. Akurasi jaringan syaraf tiruan dipengaruhi oleh jumlah lapisan tersembunyi dan neuron didalamnya. Salah satu teori yang membahas tentang jumlah lapisan tersembunyi dan neuron didalamnya adalah Teori Masters. Teori Masters merumuskan jumlah neuron berdasarkan aturan geometric pyramid. Selain itu, Teori Masters juga mengungkapkan bahwa tidak ada alasan menggunakan lebih dari dua lapisan tersembunyi. Penelitian ini bertujuan untuk membuktikan kebenaran Teori Masters menggunakan metode desain faktorial. Kombinasi yang digunakan ialah 1, 5, 10, dan 15 neuron tersembunyi. Hasil penelitian menggunakan metode desain faktorial, menunjukkan bahwa rumus teori Masters memiliki peringkat yang cukup baik yaitu 50% teratas untuk data training maupun testing. Aturan geometric pyramid memiliki akurasi yang baik pada data training. Akan tetapi aturan tersebut tidak berlaku pada data testing. Model jaringan syaraf tiruan dengan empat lapisan tersembuyi memiliki nilai akurasi RMSE (Root Mean Square Error) terbaik pada data training dan testing. Semakin banyak lapisan tersembunyi maka semakin baik nilai RMSE data training maupun data testing. Dengan demikian dapat disimpulkan bahwa Teori Masters yang menyebutkan bahwa tidak ada alasan menggunakan lebih dari dua lapisan tersembunyi, terbukti tidak valid. AbstractArtificial neural networks is a forecasting method a very common method for forecasting. Accuracy of artificial neural networks is influenced by the number of hidden layers and neurons in them. One theory that discusses the number of hidden layers and neurons in them is the Masters Theory. Masters Theory formulates the number of neurons based on geometric pyramid rules. In addition, the Masters Theory also reveals that there is no reason to use more than two hidden layers. This study aims to prove the Masters Theory using factorial design methods. The combinations used are 1, 5, 10, and 15 hidden neurons. Based on factorial design methods in this study, it can be concluded that the formula for many neurons has adequate rating of 50% above, both training and testing data. Tthe geometric pyramid rules have good accuracy in training data. However, this rule does not apply to data testing. The artificial neural network model with four hidden layers has the best RMSE (Root Mean Square Error) accuracy values in training and testing data. The more hidden layers will obtain better RMSE in both training dan testing datasets. Thus, the Masters Theory which states that there is no reason to use more than two hidden layers, proved to be invalid.
Sentiment Analysis of NU Online Applications Using Artificial Neural Network Lusia, Dwi Ayu; Anuraga, Gangga; Rahman, Fathur
Southeast Asian Journal of Islamic Education Vol 6 No 2 (2024): Southeast Asian Journal of Islamic Education, June 2024
Publisher : Faculty of Education and Teacher Training of UINSI Samarinda

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21093/sajie.v6i2.8822

Abstract

The NU Online app on the Playstore serves the needs of Muslims, especially those in Islamic boarding schools, by providing information and services. Its success is gauged not just by the number of downloads or popularity but by the quality of user interactions and how well it meets user needs. Sentiment analysis of user reviews provides deeper insights into these aspects. This research focused on finding words influencing sentiment from NU online and producing the best performance of artificial neural networks. This study collected user reviews from the NU Online app between February 9, 2021, and May 31, 2024, totalling 12613 reviews. After preprocessing, 8546 reviews remained. Using the Indonesian Sentiment Lexicon (INSET), 66% of the reviews showed positive sentiment, 21% were neutral, and 13% were negative. The words "aplikasi" (application) and "nya" (its) appeared in the top three across all sentiment classes, while "fitur" (feature) was common in both positive and negative sentiments. For neutral sentiments, "nan" was frequently mentioned. The data were split into training and testing sets in an 80:20 ratio, preserving the proportions of each sentiment class. Sentiment analysis was performed using a neural network, with input neurons ranging from the top 10 words from each sentiment class to all words. Accuracy improved as more words were used, peaking at 0.95 for the top 1690 words, compared to 0.71 for the top 10 words. The findings highlight the importance of using a comprehensive set of words to train the ANN. Including more words significantly enhances the model's performance, indicating that a richer vocabulary captures sentiment nuances better.
Educational Workshop Berbasis HOTS: Upaya Meningkatkan Kualitas Guru SMP dan SMA pada Olimpiade Guru Nasional Fernandes, Adji Achmad Rinaldo; Lusia, Dwi Ayu; Nisa, Hilwin; Hidayatulloh, Moh Zhafran; Rizqia, Anggun Fadhila; Nasywa, Alfiyah Hanun; Putri, Nazwa Anindya; Amirullah, Khoirul Insan
Seminar Nasional Penelitian dan Pengabdian Kepada Masyarakat 2025 Prosiding Seminar Nasional Penelitian dan Pengabdian Kepada Masyarakat (SNPPKM 2025)
Publisher : Universitas Harapan Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35960/snppkm.v4i1.1411

Abstract

The National Teacher Olympiad (OGN) is a prestigious event that aims to improve teacher competence, especially in the field of mathematics, through mastery of pedagogy, learning innovation, and the application of Higher Order Thinking Skills (HOTS). However, junior and senior high school teachers in Malang Regency still face obstacles in the form of limited access to training, lack of professional community, and low literacy in learning technology. This service program was carried out at PP & SMA Sumber Putih, Malang Regency, with the aim of strengthening teacher competence through strategies to strengthen positive mindsets, increase motivation, and interactive training based on Higher Order Thinking Skills. The implementation method includes educational workshops, motivational sessions, group discussions, preparation of learning modules, and reflection to measure the effectiveness of the program. The results of the activity showed an increase in teachers' mental readiness in facing the National Teachers' Olympiad, strengthening the understanding of Higher Order Thinking Skills in mathematics learning, and improving technological skills in the learning process. In addition, training modules are arranged as outputs that can be used continuously. This program contributes to improving the professionalism of teachers, encouraging participation in OGN, and building an innovative and competitive education ecosystem in Malang Regency.
Co-Authors Achmad Efendi Amirullah, Khoirul Insan Awalludiyah Ambarwati Eni Sumarminingsih Fathur Rahman Fernandes, Adji Achmad Rinaldo Gangga Anuraga Heni Kusdarwati Hidayatulloh, Moh Zhafran Nasywa, Alfiyah Hanun Nisa, Hilwin Putri, Nazwa Anindya Rizqia, Anggun Fadhila Salsabila, Imelda Semathea, Karen Simanjuntak, Yessica Maretha Suci Astutik