Agus Nursikuwagus

Mr. Agus Nursikuwagus

Universitas Komputer Indonesia, Bandung, Jawa Barat, Indonesia
+1
orcid

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 March 2017 in TELKOMNIKA (Telecommunication Computing Electronics and Control)
Verified
Journal Article
Open access
PDF file
Agus NursikuwagusAgus Nursikuwagus
Agus Nursikuwagus
ABAgis Baswara
Agis Baswara
This paper presented about study that have been created a new student choosing system by using fuzzy mamdani inference systems method. Fuzzy mamdani is used because it has characteristics such as human perceptions on choosing of students with some specified criteria. The choosing studens who want entry to the school have been difficult if it is manually process. With the fuzzy mamdani, the process can be possible completed execute and can be reduced the time of choose. To accomplish the process, the fuzzy variable is created by the national final exam scores, report grade, general competency test, physical test, interview and psychological test. Based on testing 270 data, the fuzzy mamdani has been reached 75.63% accuracy.

22 October 2022 in Journal of Imaging
Verified
Journal Article
Open access
PDF file
Agus NursikuwagusAgus Nursikuwagus
Agus Nursikuwagus
RMRinaldi Munir
Rinaldi Munir
MKMasayu Leylia Khodra
Masayu Leylia Khodra
Captioning is the process of assembling a description for an image. Previous research on captioning has usually focused on foreground objects. In captioning concepts, there are two main objects for discussion: background object and foreground object. In contrast to the previous image-captioning research, generating captions from the geological images of rocks is more focused on the background of the images. This study proposed image captioning using a convolutional neural network, long short-term memory, and word2vec to generate words from the image. The proposed model was constructed by a convolutional neural network (CNN), long short-term memory (LSTM), and word2vec and gave a dense output of 256 units. To make it properly grammatical, a sequence of predicted words was reconstructed into a sentence by the beam search algorithm with K = 3. An evaluation of the pre-trained baseline model VGG16 and our proposed CNN-A, CNN-B, CNN-C, and CNN-D models used BLEU score methods for the N-gram. The BLEU scores achieved for BLEU-1 using these models were 0.5515, 0.6463, 0.7012, 0.7620, and 0.5620, respectively. BLEU-2 showed scores of 0.6048, 0.6507, 0.7083, 0.8756, and 0.6578, respectively. BLEU-3 performed with scores of 0.6414, 0.6892, 0.7312, 0.8861, and 0.7307, respectively. Finally, BLEU-4 had scores of 0.6526, 0.6504, 0.7345, 0.8250, and 0.7537, respectively. Our CNN-C model outperformed the other models, especially the baseline model. Furthermore, there are several future challenges in studying captions, such as geological sentence structure, geological sentence phrase, and constructing words by a geological tagger.

1 September 2018 in Jurnal Teknologi dan Informasi
Verified
Journal Article
Open access
PDF file
RSRangga Sidik
Rangga Sidik
MFMia Fitriawati
Mia Fitriawati
SMSyahrul Mauluddin
Syahrul Mauluddin
Agus NursikuwagusAgus Nursikuwagus
Agus Nursikuwagus
Penjadwalan kuliah merupakan proses yang sangat kompleks, karena harus mengatur beberapa komponen seperti dosen, matakuliah, kelas, ruang dan waktu dengan memperhatikan sejumlah batasan dan syarat (constraint) tertentu. Dalam proses pembuatan jadwal kuliah ini setiap perguruan tinggi masing masing berbeda syarat dan batasan. Hal ini lah yang menyebabkan banyak dilakukan penelitian terkait optimasi penjadwalan kuliah. Perlunya optimasi penjadwalan kuliah dirasakan juga oleh program studi manajemen yang merupakan salah satu program studi di Universitas Komputer Indonesia dengan jumlah mahasiswa yang banyak. Dalam proses penjadwalan kuliah masih dilakukan secara konvensional sehingga sering terjadi bentrok jadwal dan prosesnya membutuhkan waktu yang lama. Dalam penggambaran sistem ini menggunakan metode pendekatan berorientasi objek dengan alat bantu pemodelan UML (Unified Modeling Language) dan menggunakan metode pengembangan sistem model prototipe sebagai acuan tahapan penelitian. Penelitian ini merupakan tahap awal dalam membangun sistem informasi penjadwalan kuliah. Hasil penelitian ini berupa model penerapan algoritma Ant Colony Optimization (ACO) untuk optimasi sistem informasi penjadwalan kuliah.

26 September 2018 in IOP Conference Series: Materials Science and Engineering
Verified
Conference Paper
Open access
PDF file
LML Melian
L Melian
Agus NursikuwagusAgus Nursikuwagus
Agus Nursikuwagus
This research is proposed about decision in eligibility for accepting and rejecting student in vocational school. Prediction is one area in data mining that discuss about classification. One method of algorithm in prediction is Naïve-Byes Decisions Algorithm. Variable will be counted by Naive-Byes Decisions algorithm that content of final exam, report, psychology, interview, and competency. Execution for the first step is transforming each variable into ranging values. Step forward is continued to calculate possibility value for every variable. In 270 student, the research has been resulted the validation test is 199 same as actual condition. Precision is 96,1%, recall is 99,3%. The result is meaning that the algorithm has ability to get the decision in accepting manner with 74,87% accuracy. The value gives the reasoning that has not all submitting students are accepted by the school. Meanwhile, the algorithm has giving fair process for accepting the candidate. The impact of this research is the school can predict how many students is accepted and rejected. So, school will prepare for how many rooms should be available.

15 September 2024 in JSiI (Jurnal Sistem Informasi)
Verified
Journal Article
Open access
PDF file
Agus NursikuwagusAgus Nursikuwagus
Agus Nursikuwagus
SUSuherman
Suherman
HGHendry Gunawan
Hendry Gunawan
IAIlham Alamsyah
Ilham Alamsyah
Kasus angka kematian yang terjadi di Indonesia dapat di kelompokan dalam beberapa kategori seperti natural disaster, nonnatural disaster, dan social disaster. Pemisahan suatu instans pada dataset sering menjadi hambatan Ketika melibatkan instans yang banyak. Penemuan karakteristik yang serupa akan menjadi tantangan untuk mendapatkan kluster terbaik. Penentuan jumlah kluster yang efektif terhadap dataset yang dimiliki menjadi permasalahan lain Ketika melakukan proses kluster. Berdasarkan permasalahan dan tantangan yang diperoleh, maka untuk menjawab hal ini dilakukanlah pemodelan clustering dengan bantuan algoritma clustering. Metode yang digunakan pada pengklusteran ini adalah K-Means. Metode ini telah menjadi usulan dari berbagai penelitian yang menyatakan sukses dalam melakukan clustring. Penentuan K terbaik yaitu dengan bantuan elbow curve, dengan melihat titik elbow pada hasil generasi kurva dari dataset. Rangkaian penyelesaian penelitian ini adalah dengan mengikuti flow of process datamining yang dimulai dengan Data Preprocessing, Data modeliing, dan visualization hasil. bertujuan untuk mengetahui klusterisasi penyebab kematian di Indonesia berdasarkan kategori yang di sebutkan di atas. Dataset yang digunakan adalah sebanyak 648 instans yang diambil dari rentang 2000 – 2020 mengenai kasus kematian pada 34 provinsi di Indonesia. Data preprocessing adalah melakukan cleansing data, pembersihan outlier, missing value, data transformation. Pembersihan outlier yaitu menggunakan bantuan Box Plot, sedangkan transformation menggunakan fungsi transormasi data diskrit menjadi data numerik. Pada data modelling, algoritma K-means dengan K atau banyaknya diperoleh dari hasil Elbow Curve. Selain proses clustering, penggalian pola juga dilakukan dengan metode classification yang hasilnya ditunjukan dengan akurasi sebesar 63%. Meninjau dari hasil classification, bahwa klasifikasi kematian yang berasal dari sumber sosial, tidak dapat diprediksi dengan akurat. Klasifikasi sumber kematian dari Sosial tidak berhasil dipolakan oleh mesin learning. Matrik konfusi menunjukkan hanya 55 instans yang benar untuk bencana alam, bencana non alam dan penyakit sebesar 353 yang benar, dan untuk bencana sosial tidak berhasil diprediksi. Dari hasil ini, maka dapat diperoleh tantangan baru yaitu memperbaiki akurasi dengan mempertimbangan Imbalance Class, dan Resampling yang belum digunakan pada penelitian ini. Kata kunci: Unsupervise, clustering, K-Means, euclidean distance, elbow curve,.

16 October 2024 in IEEE Access
Verified
Journal Article
Open access
Agus NursikuwagusAgus Nursikuwagus
Agus Nursikuwagus
RMRinaldi Munir
Rinaldi Munir
MKMasayu L. Khodra
Masayu L. Khodra
Deshinta Arrova DewiDeshinta Arrova Dewi
Deshinta Arrova Dewi
Image captioning is a hot topic that combines a multidiscipline task between computer vision and natural language processing. One of the tasks in the geological field is to make descriptions from the images of geological rocks. The task of a geologist is to write a content description of an image and display it as text that can be used in the future. Interpretation of the object is one of the objectives of the research, which is to traverse the image structures in depth. Shapes, colors, and structures are to be focused on to get the image’s features. The problem faced is how the separable neural network (SNN) and long short-term memory (LSTM) have an impact on the caption that can meet the geologist’s description. SNN is called Visual Attention (VaT), and LSTM is called Semantic Attention (SemAtt) as an architecture of image captioning. The result of the experiment confirms that the accuracy model for captioning gets BLEU-1 = 0.908, BLEU-2 = 0.877, BLEU-3 = 0.750, and BLEU-4 = 0.510. The evaluation score is compared to those of other evaluators, such as Meteor and RougeL, which get 0.670 and 0.623, respectively. The model confirms that it outperforms the baseline model. Referring to the evaluations, we concluded that the model was able to generate captioned geological rock images that met the geologist’s description. Precision and recall have supported the models in providing the predicted word that is suitable for the image features.