cover
Contact Name
I Ketut Resika Arthana, S.T., M.Kom
Contact Email
resika@undiksha.ac.id
Phone
-
Journal Mail Official
-
Editorial Address
-
Location
Kota denpasar,
Bali
INDONESIA
Jurnal Sains dan Teknologi
ISSN : 23033142     EISSN : 25488570     DOI : -
Core Subject : Science, Education,
Jurnal Sains dan Teknologi(JST) is a journal aims to be a peer-reviewed platform and an authoritative source of information. We publish original research papers, review articles and case studies focused on Mathematic, Biology, Physic, Chemistry, Informatic, Electronic and Machine as well as related topics. All papers are peer-reviewed by at least two referees. JST is managed to be issued twice in every volume.
Arjuna Subject : -
Articles 648 Documents
Perbandingan Metode Transfer Learning untuk Identifikasi Tumbuhan Herbal Berbasis Lontar Usada Taru Pramana Desak Made Sidantya Amanda Putri; G K Gandhiadi; I GN Lanang Wijayakusuma
JST (Jurnal Sains dan Teknologi) Vol. 14 No. 1 (2025): April
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/jstundiksha.v14i1.92414

Abstract

Indonesia dikenal memiliki keanekaragaman hayati yang melimpah, termasuk tumbuhan herbal yang dimanfaatkan dalam pengobatan tradisional, seperti di Bali yang beberapa masyarakatnya masih menggunakan tumbuhan herbal sebagai bahan pengobatan.  Namun, identifikasi dan klasifikasi tumbuhan herbal masih menjadi tantangan karena kesamaan morfologi antarspesies, yang dapat memengaruhi efektivitas pengobatan. Oleh karena itu, penelitian ini menganalisis dan mengembangkan metode klasifikasi tumbuhan herbal dengan pendekatan transfer learning menggunakan arsitektur Convolutional Neural Network (CNN), yaitu MobileNet-V2 dan ResNet-50 V2, guna meningkatkan akurasi klasifikasi. Penelitian ini menggunakan dataset TPHerbleaf yang berisi 1000 citra dari 50 jenis daun tumbuhan herbal yang tercatat dalam Lontar Usada Taru Pramana. Data diproses dengan teknik augmentasi dan fine-tuning, kemudian model diuji dengan membandingkan arsitektur usulan dengan penelitian terdahulu. Hasilnya, MobileNet-V2 mencapai akurasi 98,65%, sedangkan ResNet-50 V2 mencapai 98,48%, menunjukkan peningkatan signifikan dibanding model sebelumnya. MobileNet-V2 lebih efisien dalam penggunaan sumber daya, sementara ResNet-50 V2 lebih stabil dalam pelatihan jaringan yang dalam. Penelitian ini berkontribusi pada pengembangan metode klasifikasi tanaman herbal berbasis CNN yang lebih akurat dan aplikatif, serta dapat digunakan untuk mendukung pemanfaatan tanaman herbal secara lebih luas.
Pengaruh Shear Force terhadap Percepatan Pembentukan Granular Indigenous Microalgal-Bacterial Consortium dari Palm Oil Mill Effluent Muhammad Faisal Dharma; Elystia, Shinta; David Andrio
JST (Jurnal Sains dan Teknologi) Vol. 14 No. 1 (2025): April
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/jstundiksha.v14i1.92583

Abstract

Produksi Crude Palm Oil (CPO) yang besar akan menyebabkan Palm Oil Mill Effluent (POME) yang dihasilkan semakin besar. POME mengandung zat organik yang tinggi dan kaya akan nutrisi N, P, dan K yang dapat dimanfaatkan oleh mikroalga dan bakteri. Granular Indigenous Microalgal-Bacterial Consortium (G-IMBC) adalah salah satu alternatif pengolahan POME dengan memanfaatkan mikroalga dan bakteri. G-IMBC memiliki kelebihan dibandingkan dengan pengolahan alternatif lainnya karena mampu mendegradasi polutan dengan beban COD dan nitrogen total yang tinggi secara simultan dalam waktu tinggal hidraulik yang singkat. Penelitian ini bertujuan untuk menganalisis pengaruh shear force terhadap karakteristik dan kinerja G-IMBC dalam penyisihan nutrisi pada POME. Mikroalga dan bakteri yang digunakan berasal dari POME pada kolam fakultatif yang telah dilakukan pretreatment terlebih dahulu. Mikroalga dan biomassa bakteri indigenous diinokulasikan ke dalam fotobioreaktor dan dioperasilan dengan shear force yang berbeda yaitu PBR 1 (1,7 cm/s), PBR 2 (2,5 cm/s), dan PBR 3 (3,4 cm/s). Karakteristik G-IMBC terbaik dihasilkan pada PBR 3, yaitu pada hari ke-18 dengan diameter granular 0,8-1 mm, VSS sebesar 11,88 g/L, SVI5 sebesar 32,7 mL/g, densitas sebesar 3,1 g/mL, serta efisiensi penyisihan COD dan nitrogen total yang tinggi, yaitu 95,8% dan 96,7%. Hal ini terjadi karena peningkatan shear force dalam proses pembentukan G-IMBC mampu memperkuat struktur G-IMBC, sehingga efisiensi dalam penyisihan substrat menjadi lebih optimal.
Perbandingan Metode LSTM dan TCN untuk Prediksi Gelombang Laut Berdasarkan Enam Parameter Oseanografi Ni Nyoman Bintang Marscelina; I GN Lanang Wijayakusuma; Putu Veri Swastika
JST (Jurnal Sains dan Teknologi) Vol. 14 No. 1 (2025): April
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/jstundiksha.v14i1.92590

Abstract

Perubahan kondisi oseanografi seperti variabilitas gelombang laut mengancam keselamatan pengguna pantai dan aktivitas maritim di Pantai Mooloolaba, Australia. Penelitian ini dilatarbelakangi oleh kebutuhan mendesak untuk mengembangkan model prediksi yang mampu menangkap pola temporal jangka panjang dari parameter oseanografi secara akurat. Oleh karena itu, penelitian ini membandingkan dua pendekatan deep learning, yaitu long short-term memory (LSTM) dan temporal convolutional network (TCN), guna mengoptimalkan prediksi perilaku gelombang laut berdasarkan enam parameter oseanografi. Menggunakan metode kuantitatif dengan desain komparatif eksperimental, penelitian ini memanfaatkan data sekunder dari Queensland Government Data dengan interval pengukuran 30 menit (20 April 2000 – 31 Agustus 2024). Setelah pra-pemrosesan, data dibagi menjadi 80% pelatihan dan 20% pengujian. Hasil evaluasi menunjukkan bahwa TCN memiliki nilai RMSE lebih rendah dibandingkan LSTM pada semua parameter, baik pada data latih maupun uji. Oleh karena itu, TCN lebih unggul dalam menangkap pola temporal jangka panjang dan lebih efektif untuk mitigasi risiko kondisi laut ekstrem.
Integrasi Komponen Elektonika Berbasis ESP32 dan Sensor Kelembapan untuk Penyiraman Otomatis Pada Tanaman Anggrek Afdal Bintang Syahputra; Agus Ulinuha
JST (Jurnal Sains dan Teknologi) Vol. 14 No. 1 (2025): April
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/jstundiksha.v14i1.93007

Abstract

Tanaman anggrek merupakan salah satu tanaman hias yang ada di Indonesia dengan nilai estetika tinggi. Waktu proses pertumbuhan dan perkembangan tanaman akan membutuhkan sensor kelembaban tanah yang sangat optimal agar selalu seimbang. Keadaan tanah yang memiliki kadar air yang berlebihan akan menyebabkan pembusukan di bagian akar tumbuhan, sedangkan jika tanaman kekurangan kadar air juga membuat tanaman tidak dapat berbungan. Oleh karena itu, saat ini sudah ada pengembangan sistem penyiraman otomatis yang sudah menjadi topik penting di era modern ini. Penelitian ini bertujuan untuk menganalisis sistem penyiraman otomatis pada tanaman anggrek dengan menggunakan elektonika berbasis ESP32 dan sensor kelembaban yang menjadi mikrokontorel sekaligus konfigurasi terhadap aplikasi Bly melalui IoT. Metode penelitian yang digunakan adalah campuran yaitu dengan melibatkan 15 partisipan sebagai subjek penelitian, terdiri dari 10 orang yang diuji di dalam ruangan dan 10 orang di luar ruangan. Sistem ini dirancang untuk mendeteksi keembaban tanah dan mengatur penyiraman secara otomatis untuk menjaga kesehatan tanaman anggrek. Hasil penelitian menunjukkan sistem otomatis kelembaban tanah yang telah dirancang dengan menggunakan sensor capacitive soil moisture SKU:SEN0193, relay 5 volt, pompa air mini dc 12 volt, dan mikrokontroler NodeMCU ESP32 berbasis IoT hingga memperoleh hasil yang baik. Penelitian ini sudah melalui pengujian laboratorium dan lapangan, sistem ini juga sudah terbukti dengan respon lingkungan dalam mengoptimalkan penggunaan air berdasarkan kebutuhan tanaman.
Aktivitas Antibakteri Ekstrak Daun Rivina humilis L. Terhadap Staphylococcus aureus dan Pseudomonas aeruginosa Yuliani, Ririn Kholifatu; Ilma, Nur Malika; Tufahati, Nadhira; Salimah, Azizah Zahratus; Aji, Oktira Roka
JST (Jurnal Sains dan Teknologi) Vol. 14 No. 1 (2025): April
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/jstundiksha.v14i1.93228

Abstract

Rivina humilis (getih-getihan) is a wild plant that contains bioactive compounds that have the can to be used as antibiotics. This study aims to determine the antibacterial activity of ethanol extract of getih-getihan leaves against Staphylococcus aureus (Gram positive) and Pseudomonas aeruginosa (Gram negative) bacteria through in vitro experimental research. The subject used in the study was getih-getihan (Rivina humilis) leaves which were fresh and good without any disease. Data collection methods were carried out through laboratory experiments including extraction, metabolic analysis using LC-HRMS, antibacterial testing using disc diffusion, MIC and MBC. Data were analyzed using ANOVA and Duncan's post-hoc test with a significant value of p<0.05. The results of ethanol extract of getih-getihan leaves showed flavonoids, phenols, triterpenoids, amino acids, fatty acids, and carboxylic acids. The extract is can to inhibit and kill both bacteria, which is indicated by the presence of an inhibition zone, a decrease in absorbance, and no colony growth on the test media. In conclusion, ethanol extract of getih-getihan leaves has antibacterial activity against S. aureus and P. aeruginosa. The implication is that ethanol extract of getih-getihan leaves has the potential to be developed as a natural antibacterial agent that supports alternative treatment against bacterial infections against S. aureus and P. aeruginosa.
Desain Arsitektur Enterprise Berbasis Risiko Teknologi Informasi untuk Sistem Pelatihan: Integrasi COBIT 2019 dan TOGAF ADM Desyaf Putra; Dinar Mutiara Kusumo Nugraheni; Jatmiko Endro Suseno
JST (Jurnal Sains dan Teknologi) Vol. 14 No. 1 (2025): April
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/jstundiksha.v14i1.93498

Abstract

The increasing utilization of Information Technology (IT) in training implementation at the Agricultural Training Center (Bapeltan) Jambi presents challenges in IT risk management, system effectiveness, and infrastructure resilience. This study aims to identify the key IT risks faced by Bapeltan Jambi and design an Enterprise Architecture (EA)-based solution using TOGAF ADM with a risk analysis approach from COBIT 2019. The research methods employed include observations, interviews with various stakeholders, and risk analysis using APO12 (Manage Risk) in COBIT 2019. The research findings indicate that IT infrastructure disruptions, limited network capacity, data security vulnerabilities, and lack of system integration are the main factors affecting the effectiveness of training implementation. A total of 25 risk scenarios have been identified and categorized in the Risk Register, covering likelihood levels, impact, and mitigation strategies. Based on the analysis results, an enterprise architecture design based on TOGAF ADM has been developed with the models of Architecture Vision, Business Architecture, Information Systems Architecture, Technology Architecture, and Migration Planning to enhance the performance of IT-based training systems.
A Machine Learning Framework for Automatic Speech Transcription and Summarization Using HMM and TextRank Kurnia , Yusuf; Kristen; Rossi , Ardiane; Junaedi; Hermawan , Aditiya
JST (Jurnal Sains dan Teknologi) Vol. 14 No. 1 (2025): April
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/jstundiksha.v14i1.94184

Abstract

This study is motivated by the increasing need to process audio data efficiently, such as in meetings, lectures, and interviews, which are usually still done manually. This manual process is time-consuming and prone to human error, so an automated system is needed that can convert speech into text and summarize information accurately. The main objective of this study is to develop an automated system that integrates the Hidden Markov Model (HMM) for speech transcription and TextRank for text summarization, and to evaluate the performance of the system. This study uses a quantitative experimental approach with research subjects in the form of audio data in MP3 format obtained from various activities, such as meetings, lectures, and interviews. The audio data is processed using the feature extraction method using Mel-Frequency Cepstral Coefficients (MFCC), then transcribed using HMM and summarized using the TextRank algorithm. Data analysis is carried out by measuring the accuracy of the transcription using the Word Error Rate (WER) and evaluating the quality of the summary using the ROUGE metric. This system is tested on three audio categories with varying complexity. The results show that the system achieves high transcription accuracy, especially for interview audio (WER: 7.6%) and effective summarization performance (ROUGE-1: 0.78, ROUGE-L: 0.74). Furthermore, the automated workflow shows up to 96% time efficiency improvement compared to the manual method. These findings demonstrate the practical feasibility of combining probabilistic and graph-based algorithms to automate large-scale audio data processing. This approach significantly reduces human workload while ensuring accuracy and consistency. This research has implications for contributing to the advancement of hybrid natural language processing systems and providing a solid foundation for future integration with transformer-based abstractive summarization and multilingual scalability.
Effectiveness of Monocrystalline Solar Panel Tilt Angle to Output Power and Efficiency: Case study for Singaraja-Bali Yasmini, Luh Putu Budi; Valentina, Diah Novita; Risha, Nurfa
JST (Jurnal Sains dan Teknologi) Vol. 14 No. 1 (2025): April
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/jstundiksha.v14i1.94203

Abstract

The tilt angle orientation influences the performance of solar panels. This study aims to determine the optimal tilt angle of solar panels, especially in the Singaraja area, which has the highest efficiency output power. This type of research is experimental research. The subjects involved in this study were monocrystalline solar panels tested at various tilt angles, namely 0°, 10°, 20°, 30°, 40°, 50°, and 60°. The data collection method was carried out by directly measuring the intensity of the incident sunlight and recording the output voltage and current values of the solar panel at each tilt angle variation. The data obtained was then analyzed to determine the output power and efficiency of the solar panel at each tilt angle by comparing the measurement results at various test times so that the tilt angle that produces the best performance under local environmental conditions can be identified. The results showed that the highest output power was obtained at 12:05 a.m. with a 10° tilt angle of 37.43 Watts. The lowest output power was obtained at 9:30 a.m. with a 60° tilt angle of 9.63 Watts. The highest efficiency is produced at an inclination angle of 10° by 11.62%. The lowest efficiency was obtained at a 60° tilt angle of 10.47%. The optimal tilt angle in Singaraja is 10° to produce high output power and good efficiency. Based on these findings, it can be concluded that the tilt angle of solar panels significantly influences the output power and efficiency of monocrystalline solar panels in the Singaraja region of Bali.
Identification of Chemical Compounds of Guava Leaf Fractions using Phytochemical Test, UV-Vis Spectrophotometry and GC-MS Wardani, Lailatul Munawaroh Dewi Kusuma; Andayani, Sri; Maimunah, Yunita
JST (Jurnal Sains dan Teknologi) Vol. 14 No. 1 (2025): April
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/jstundiksha.v14i1.94481

Abstract

Fraction is the result of fractionation of the process of separating compounds based on the polarity properties of the solvent. Thin layer chromatography is the first step in fractionation analysis to determine the eluent that can produce guava leaf fractions. The guava leaf fraction contains chemical compounds that play a role in improving the immune system. Guava leaves contain flavonoid compounds, especially quercetin. The effort made to find out the content of guava leaves is to analyze further. Some analyses to determine the content of active compounds, especially in guava leaf fractions, are using phytochemical tests, UV-VIS spectrophotometry, and GC-MS. The combination of these three methods can obtain comprehensive information about the content of chemical compounds in plants, from the identification of compound groups to the characteristics of specific compounds. Phytochemical analysis is an initial analytical method carried out to examine the content of chemical compounds in medicinal plants. Ultra Violet-Visble (UV-Vis) spectrophotometer is one of the chemical analysis methods to determine the composition of a sample, both quantitatively and qualitatively, based on the interaction between matter and light. The Gas Cromatography-Mass Spectrometry (GC-MS) method is a method of separating samples using gas chromatography while analyzing the compounds using mass spectroscopy. The purpose of this study is to analyze the active compounds contained in the guava leaf fraction. This research uses a qualitative method by collecting data by observation and literature study.  The results showed that the results of the analysis of guava leaf fractions using phytochemical tests, UV-VIS spectrophotometers, and GC-MS, showed that the fraction contained terpenoid and flavonoid group compounds.
The Effectiveness of Technology Based Internship Programs in Improving the Skills of Industrial Technology Vocational School Students Widjaja, Warkianto
JST (Jurnal Sains dan Teknologi) Vol. 14 No. 2 (2025): July
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/jst-undiksha.v14i2.96975

Abstract

The lack of human resource capabilities in industrial technology has an impact on the lack of development of real industrial needs. The urgency of researching the need for human resources in facing advanced industrial technology projects. Therefore, this study aims to evaluate the factors that influence the obstacles and difficulties of technology-based industrial internship programs in implementing practical practices in the field. The method used is a quantitative survey type. The sample amounted to 450 industrial engineering students who were selected randomly. The data collection technique was done through an online questionnaire based on a Likert scale of 36 items. The data analysis technique used descriptive statistics, assisted by SPSS Version 29.0. With the DIF test, factor analysis, and reliability. The study's results indicate that obstacles and difficulties in program implementation are inadequate preparation, minimal industry involvement, and inadequate quality of project internships. However, programs with technology assistance were effective in improving future skills. Technology-based internship programs can address competency gaps by providing relevant hands-on experience recommendations, increasing industry collaboration, and supporting training in using technology as practical steps. The implications of this study indicate that technology integration in internship activities can be a significant strategy to bridge the gap between student competencies and industry needs.