Universitas Putera Batam (UPB): Open Journal Systems
Not a member yet
3910 research outputs found
Sort by
ANALISIS KLASIFIKASI ULASAN APLIKASI SIREKAP 2024 MENGGUNAKAN EKSTRAKSI FITUR DISTILBERT DAN METODE SUPPORT VECTOR MACHINE
Full text linkThis study aims to classify reviews of the SIREKAP 2024 application automatically using the DistilBERT feature extraction method and the Support Vector Machine (SVM) algorithm. The data used includes 8,538 user reviews from the Google Play Store with five Rating categories as the target variable. After undergoing 10-Fold cross-validation, the average F1-Score obtained was 36.62%, with the highest performance reaching 37.16%. The analysis indicates that data imbalance is the main obstacle in improving the model's accuracy, particularly in the minority class. The study concludes that the combination of DistilBERT and SVM yields suboptimal results and requires further optimization. Recommendations are provided to improve model accuracy and enhance the quality of the application based on user reviews.Penelitian ini bertujuan untuk mengklasifikasikan ulasan aplikasi SIREKAP 2024 secara otomatis menggunakan metode ekstraksi fitur DistilBERT dan algoritma Support Vector Machine (SVM). Data yang digunakan mencakup 8.538 ulasan pengguna dari Google Play Store dengan lima kategori Rating sebagai variabel target. Setelah melalui proses validasi silang dengan 10 Fold, rata-rata nilai F1-Score yang diperoleh adalah 36,62%, dengan performa tertinggi mencapai 37,16%. Analisis menunjukkan bahwa ketidakseimbangan data menjadi kendala utama dalam meningkatkan akurasi model, khususnya pada kelas minoritas. Penelitian ini menyimpulkan bahwa kombinasi DistilBERT dan SVM memberikan hasil yang tidak cukup baik dan masih membutuhkan optimasi lebih lanjut. Rekomendasi diberikan untuk meningkatkan akurasi model dan memperbaiki kualitas aplikasi berdasarkan ulasan pengguna
PERANCANGAN SISTEM PEMBELAJARAN LMS MATAKULIAH BAHASA INDONESIA: PERANCANGAN SISTEM PEMBELAJARAN LMS MATAKULIAH BAHASA INDONESIA
No full textThe flexibility of the concept of education is slowly shifting from face-to-face learning, which means that the presence of educators and students is the main indicator in the teaching and learning process in the classroom to a distance education system. Learning Management System (LMS) is a web-based software program that is widely used for management activities, documentation, monitoring, reporting, administration and distribution of educational content, training programs, technical manuals, instructional videos, digital libraries, and learning projects. The objectives of this study are: 1) To find out how the workflow of the Indonesian language course learning system works; 2) To determine the model and function of the system in detail; and 3) To find out the data flow through a visual approach. The research method used to complete this research is the iterative and agile systems development lifecycle method. from the results of the study it can be seen that there are 3 entities that play a role in the process of this system, namely: Admin, Lecturer, and StudentFleksibelitas konsep pendidikan secara berlahan lahan mulai bergesar dari pembelajaran tatap muka, yang berarti kehadiran pendidik dan peserta didik menjadi indikator utama dalam proses belajar mengajar dikelas ke sistem pendidikan jarak jauh. Learning Management System (LMS) merupakan program perangkat lunak berbasis web yang banyak dipakai untuk aktivitas manajemen, dokumentasi, pemantauan, pelaporan, administrasi dan distribusi konten pendidikan, program pelatihan, manual teknis, video instruksional, perpustakaan digital, dan proyek pembelajaran. Tujuan dari penelitian ini adalah: 1)Untuk mengetahui bagaimana alur kerja sistem pembelajaran matakuliah Bahasa Indonesia; 2) Untuk menetukan model dan fungsi sistem secara mendetail; dan 3) Mengetahui alur data melalui pendekatan visual. Metode penelitian yang digunakan untuk menyelesaikan penelitian ini adalah metode iterative and agile systems development lifecycle. dari hasil penelitian dapat diketahui bahwasanya terdapat 3 entitas yang berperan dalam prosen sistem ini, yaitu: Admin, Dosen, dan Mahasisw
ANALISIS TINGKAT KEPUASAN PELANGGAN TERHADAP PELAYANAN CONTACT CENTER PLN 123 MENGGUNAKAN METODE PIECES
No full textAs the demands for quality service in the digital era increase, strategic companies such as PT. PLN (Persero) utilize Contact Center 123 as the front line of communication with customers. The background of the success of other digital innovations such as the PLN Mobile application which has been proven to significantly increase customer satisfaction, emphasizes the urgency of continuous evaluation of all service channels. This study aims to measure the level of customer satisfaction with the PLN 123 Contact Center service and provide strategic recommendations for improving quality in the future. The method used is a comprehensive analysis with the PIECES framework, which evaluates six dimensions: Performance, Information, Economy, Control, Efficiency, and Service. The results of the study showed very satisfactory system performance, with all dimensions achieving a level of conformity above 90%. The Control aspect obtained the highest score (94.81%), while the Efficiency aspect was recorded as the lowest (90.26%). Although the overall performance was very good, these findings indicate that there is clear room for improvement in the efficiency area to improve the customer experience.As the demands for quality service in the digital era increase, strategic companies such as PT. PLN (Persero) utilize Contact Center 123 as the front line of communication with customers. The background of the success of other digital innovations such as the PLN Mobile application which has been proven to significantly increase customer satisfaction, emphasizes the urgency of continuous evaluation of all service channels. This study aims to measure the level of customer satisfaction with the PLN 123 Contact Center service and provide strategic recommendations for improving quality in the future. The method used is a comprehensive analysis with the PIECES framework, which evaluates six dimensions: Performance, Information, Economy, Control, Efficiency, and Service. The results of the study showed very satisfactory system performance, with all dimensions achieving a level of conformity above 90%. The Control aspect obtained the highest score (94.81%), while the Efficiency aspect was recorded as the lowest (90.26%). Although the overall performance was very good, these findings indicate that there is clear room for improvement in the efficiency area to improve the customer experience
ANALISIS SENTIMEN PENGGUNA APLIKASI MYPERTAMINA MENGGUNAKAN METODE NAÏVE BAYES BERBASIS DATA ULASAN DI PLAY STORE
No full textMyPertamina is PT Pertamina's digital application designed to support cashless and targeted distribution of subsidized fuel. Although it has been downloaded by millions of users, the app has received a variety of responses recorded in the form of reviews on the Google Play Store. This research was conducted to determine the tendency of user opinions through an automated sentiment analysis approach. Review data is collected by utilizing the Google Play Scraper library, then processed through preprocessing stages such as text normalization, character cleaning, tokenization, common word removal, and stemming. Positive and negative sentiment labels are assigned with the help of the IndoBERT pre-trained model. The next process includes conversion of text to numerical form with TF-IDF method and application of Multinomial Naïve Bayes algorithm for classification. The model is tested using confusion matrix and 10-Fold Cross Validation. The results showed that the majority of user reviews were negative (63.65%), and the classification model achieved 89.25% accuracy, 87.5% precision, 82% recall, and 84.69% F1-score. This shows that the approach used is effective in identifying public perceptions of MyPertamina application services.MyPertamina merupakan aplikasi digital milik PT Pertamina yang dirancang untuk mendukung distribusi bahan bakar bersubsidi secara non-tunai dan tepat sasaran. Meskipun telah diunduh oleh jutaan pengguna, aplikasi ini menerima beragam tanggapan yang terekam dalam bentuk ulasan di Google Play Store. Penelitian ini dilakukan untuk mengetahui kecenderungan opini pengguna melalui pendekatan analisis sentimen otomatis. Pengambilan data ulasan dilakukan dengan memanfaatkan library Google Play Scraper, lalu diproses melalui tahapan preprocessing seperti normalisasi teks, pembersihan karakter, tokenisasi, penghapusan kata umum, dan stemming. Label sentimen positif dan negatif diberikan dengan bantuan model pra-latih IndoBERT. Proses selanjutnya mencakup konversi teks ke bentuk numerik dengan metode TF-IDF dan penerapan algoritma Multinomial Naïve Bayes untuk klasifikasi. Model diuji menggunakan confusion matrix dan validasi silang 10-Fold Cross Validation. Hasil penelitian menunjukkan bahwa mayoritas ulasan pengguna bersifat negatif (63,65%), dan model klasifikasi berhasil mencapai akurasi sebesar 89,25%, presisi 87,5%, recall 82%, serta F1-score sebesar 84,69%. Hal ini menunjukkan bahwa pendekatan yang digunakan efektif dalam mengidentifikasi persepsi masyarakat terhadap layanan aplikasi MyPertamina
PERANCANGAN WEBSITE PEMESANAN TRAVEL ONLINE PADA PT NETRAL LINTAS NUSANTARA
No full textAs work to the literature, it uses design and development of the long distance online web-based ticketing system on PT Netral Lintas Nusantara as inter city land transportation with based on Pekanbaru, Riau. There are problems inherent from the current manual ticketing system among which are slow ticketing process to the services, high chances of human errors to the data entry and limited reach of the customer service. The main objective of this study is to develop computerized booking and ticketing system on-line that will minimize of ticketing and booking and expose brethren in Christ, to global network, such that there would be access to airports with investment in the international and local and international airline and flight, which users can upload the softcopy of card to board their flight. Discretion is also designed to solve the existing system inadequacies through system requirement analysis, database design and website development method. The antinicipated output of the design is a remarkable performance enhancement of PT Netral Netral Lintas Nusantara's operational system, 24/7 service expansion in real time to customers, and an enormous administration manual usage reduction. In this information age, the official launching of the website is a key action in modernising services that enables the company to enhance competitiveness on the digital platform.Sebagai karya literatur, digunakan perancangan dan pengembangan sistem tiket online jarak jauh berbasis web pada PT Netral Lintas Nusantara sebagai angkutan darat antar kota yang berbasis di Pekanbaru, Riau. Permasalahan yang terdapat pada sistem tiket manual yang ada saat ini diantaranya adalah lambatnya proses tiket ke layanan, tingginya kemungkinan terjadinya kesalahan manusia (human error) pada entri data dan terbatasnya jangkauan layanan pelanggan. Tujuan utama dari studi ini adalah untuk mengembangkan sistem pemesanan dan tiket yang terkomputerisasi secara online yang akan meminimalisir pemesanan dan tiket serta memperkenalkan saudara seiman dalam Kristus, ke jaringan global, sehingga akan ada akses ke bandara dengan investasi pada maskapai penerbangan dan penerbangan internasional dan lokal dan internasional, dimana pengguna dapat mengunggah softcopy kartu untuk menaiki penerbangan mereka. Kebijaksanaan juga dirancang untuk mengatasi kekurangan sistem yang ada melalui analisis kebutuhan sistem, desain database, dan metode pengembangan situs web. Hasil yang diharapkan dari rancangan ini adalah peningkatan kinerja yang luar biasa pada sistem operasional PT Netral Netral Lintas Nusantara, perluasan layanan 24/7 secara real-time kepada pelanggan, dan pengurangan penggunaan manual administrasi secara besar-besaran. Di era informasi ini, peluncuran resmi situs web merupakan tindakan kunci dalam modernisasi layanan yang memungkinkan perusahaan meningkatkan daya saing di platform digital
SISTEM PEMANTAUAN KUALITAS AIR IKAN NILA MEDIA BIOFLOK MENGGUNAKAN ALGORITMA LSTM DAN SVM
No full textTilapia farming with biofloc systems faces challenges in maintaining water quality stability due to fluctuations in temperature and pH that can affect fish health. To overcome this, this research designs an Internet of Things (IoT)-based water quality monitoring system integrated with machine learning-based prediction and classification methods. The system is built using NodeMCU ESP8266, DS18B20 (temperature), and PH-4502C (pH) sensors. Historical data from the sensors is processed using the Long Short-Term Memory (LSTM) algorithm to predict the values of water quality parameters. Next, the predicted results are classified using a Support Vector Machine (SVM) into three categories: “Air Terlalu Asam”, “Kualitas Air Baik”, dan “Air Terlalu Basa”. The classification information is then automatically sent to the farmer via the Telegram application. Model testing showed very high performance, where the LSTM prediction model achieved an R-squared value of 0.8871 with a very low error rate (MAE: 0.0115; MSE: 0.0019; RMSE: 0.0436), while the SVM classification model managed to achieve an accuracy of 99.86%. The implementation of this system is expected to assist farmers in making quick and precise decisions, reduce the risk of fish mortality, and increase time and labor efficiency in biofloc pond operations..Tilapia farming with biofloc systems faces challenges in maintaining water quality stability due to fluctuations in temperature and pH that can affect fish health. To overcome this, this research designs an Internet of Things (IoT)-based water quality monitoring system integrated with machine learning-based prediction and classification methods. The system is built using NodeMCU ESP8266, DS18B20 (temperature), and PH-4502C (pH) sensors. Historical data from the sensors is processed using the Long Short-Term Memory (LSTM) algorithm to predict the values of water quality parameters. Next, the predicted results are classified using a Support Vector Machine (SVM) into three categories: “Air Terlalu Asam”, “Kualitas Air Baik”, dan “Air Terlalu Basa”. The classification information is then automatically sent to the farmer via the Telegram application. Model testing showed very high performance, where the LSTM prediction model achieved an R-squared value of 0.8871 with a very low error rate (MAE: 0.0115; MSE: 0.0019; RMSE: 0.0436), while the SVM classification model managed to achieve an accuracy of 99.86%. The implementation of this system is expected to assist farmers in making quick and precise decisions, reduce the risk of fish mortality, and increase time and labor efficiency in biofloc pond operations.
THE ROLE OF ASN SIMPONI ON WORKERS DISCIPLINE AND PERFORMANCE AT DEPARTMENT OF TRANSPORTATION, SOUTH EAST SULAWESI PROVINCIAL (THE INVOLVEMENT OF DIRECT SUPERVISORSHIP)
Full text linkThis study aims to look at, describe and analyze (1) the implementation of the policy for implementing the Simponi ASN Application to the Department of Transportation, Southeast Sulawesi Provincial ASN, and (2) the impact of implementing the policy for implementing the Simponi ASN Application on ASN discipline and performance with the involvement of direct supervisor supervision. This type of research is descriptive research with a qualitative approach. Research with a qualitative description method is used so that researchers can explain more deeply and comprehensively about the description of cases that occur in the field. Research has been carried out in Kendari City in December 2024-February 2025. The results of the study concluded that (1) the implementation of the Simponi ASN Application policy implementation has been running in terms of indicators of policy size and objectives, resources, characteristics of implementing agents, attitudes or tendencies of implementers, inter-organizational communication and implementing activities and the environment economic, social and political. And (2) the implementation of the Simponi ASN Application policy has an impact on the discipline and performance of ASN which is supported by the supervision of the direct supervisor.Penelitian ini bertujuan untuk melihat, mendeskripsikan dan menganalisis (1) implementasi kebijakan penerapan Aplikasi Simponi ASN pada Dinas Perhubungan ASN Provinsi Sulawesi Tenggara, dan (2) dampak penerapan kebijakan penerapan Aplikasi Simponi ASN terhadap disiplin dan kinerja ASN dengan melibatkan supervisi atasan langsung. Jenis penelitian ini adalah penelitian deskriptif dengan pendekatan kualitatif. Penelitian dengan metode deskripsi kualitatif digunakan agar peneliti dapat menjelaskan lebih dalam dan komprehensif tentang gambaran kasus yang terjadi di lapangan. Penelitian telah dilaksanakan di Kota Kendari pada bulan Desember 2024-Februari 2025. Hasil penelitian menyimpulkan bahwa (1) implementasi kebijakan penerapan Aplikasi Simponi ASN telah berjalan ditinjau dari indikator besaran dan tujuan kebijakan, sumber daya, karakteristik pelaku pelaksana, sikap atau kecenderungan pelaksana, komunikasi antar organisasi dan kegiatan pelaksana serta lingkungan ekonomi, sosial dan politik. Dan (2) implementasi kebijakan Aplikasi Simponi ASN memberikan dampak terhadap disiplin dan kinerja ASN yang didukung oleh pengawasan dari atasan langsung
THE EFFECT OF WORKING CAPITAL TURNOVER (WCT) AND TOTAL ASSETS TURNOVER (TATO) ON RETURN ON EQUITY (ROE) AT PT. INDOFOOD SUKSES MAKMUR, TBK
Full text linkThis study aims to determine and analyze Working Capital Turnover (WCT) and Total Asset Turnover (TATO) on Return On Equity (ROE) at PT. Indofood sukses maju, Tbk. This type of research is an associative type of research. The research instrument used in this study is a list of tables containing sales data, current assets, current liabilities, total assets, net profit and capital at PT. Indofood sukses maju Tbk, the type of data used is secondary data obtained from financial reports obtained from the website www.indofood.com. The population in this study is the financial report data of PT. Indofood sukses maju Tbk. for 30 years since it was listed on the Indonesia Stock Exchange in 1994-2024. With the sample used from 2014-2023. The sampling technique used is purposive sampling with the criteria for the most updated financial report data available for 10 consecutive years, namely from 2014-2023. The data collection technique in this study is documentation and literature study. The data analysis techniques used are classical assumption tests (normality test, multicollinearity test, heteroscedasticity test, autocorrelation test), multiple linear regression analysis, correlation coefficient, determination coefficient, t-test and f-test. The results of this study state that Working Capital Turnover (X1) and Total Asset Turnover (X2) have no significant effect on Return On Equity (Y) at PT. INDOFOOD SUKSES MAKMUR, tbk.Penelitian ini bertujuan untuk mengetahui dan menganalisis Total Asset Turnover (TATO) dan Working Capital Turnover (WCT) terhadap Return On Equity (ROE) pada PT. Indofood sukses makmur, Tbk. Jenis penelitian ini merupakan jenis penelitian asosiatif. Instrumen penelitian yang digunakan dalam penelitian ini adalah daftar tabel yang berisi data penjualan, aktiva lancar, utang lancar, total aset, laba bersih dan modal pada PT. Indofood sukses makmur Tbk, jenis data yang digunakan adalah data sekunder yang didapat dari laporan keuangan yang diperoleh dari website www.indofood.com. Populasi dalam penelitian ini adalah data laporan keuangan PT. Indofood sukses makmur Tbk. selama 30 tahun sejak tercatat pada bursa efek Indonesia pada tahun 1994-2024. Dengan sampel yang digunakan dari tahun 2014-2023. Dengan Tehnik sampling yang digunakan yaitu purposive sampling dengan kriteria data laporan keuangan terupdate yang tersedia selama 10 tahun berturut-turut yaitu dari tahun 2014-2023. Teknik pengumpulan data dalam penelitian ini adalah dokumentasi dan studi pustaka. Teknik analisis data yang digunakan adalah uji asumsi klasik (uji normalitas, uji multikolinearitas, uji heterokedastisitas, uji autokorelasi), analisis regresi linear berganda, koefisien korelasi, koefisien determinasi, uji t dan uji f. Hasil penelitian ini menyatakan bahwa Working Capital turnover (X1) dan Total Asset Turnover (X2) tidak berpengaruh dan signifikan terhadap Return On Equity (Y) pada PT. INDOFOOD SUKSES MAKMUR, Tbk
IMPLEMENTASI METODE RULE-BASED REASONING PADA EXPERT SYSTEM DALAM MENDETEKSI GANGGUAN PENCERNAAN PADA ANAK
Full text linkDigestive disorders in children are a serios health issue globally, with diarrhea being a leading cause of mortality among toddlers. In Batam City, the hitgh incidence of acute diarrhea and other digestive disorders in children underscores the urgency for better management. Contributing factors include immature immune systems, unhealthy eating patterns, poor sanitation, low parental understanding of Clean and Healthy Living Behavior, and limited access to medical information and facilities. This problem is futher exacerbated by overlapping symptoms of digestive disorders, making early diagnosis difficult. To address these challenges, this research aims to design and build a web-based expert system application capable of detecting digestive disorders in children. The system implements a Rule-Based Reasing method with a Forward Chaining algorithm, which logically matches user-inputted symptoms against rules in tis knowledge base to generate a rapid initial diagnosis. This application focuses on three main indicators of degetive disorders : Diarrhea, Constipation, and GERD, for children aged 5 months to 12 years. Thus, this expert system is expected to be an interactive tool that enhances parental understanding, facititates early diagnosis of digestive disorders in children, and serves ad a reference for healthcare professionals and the development of health information systems.Gangguan pencernaan pada anak merupakan masalah kesehatan serius di seluruh dunia, dengan diare menjadi penyebab utama kematian balita. Di Kota Batam, tingginya insiden diare akut dan gangguan pencernaan lainnya pada anak-anak menggarisbawahi urgensi untuk penanganan yang lebih baik. Faktor-faktor yang berkontribusi meliputi sistem kekebalan tubuh yang belum matang, pola makan yang tidak sehat, sanitasi yang buruk, rendahnya pemahaman orang tua tentang Perilaku Hidup Bersih dan Sehat, serta terbatasnya akses terhadap informasi dan fasilitas medis. Masalah ini semakin diperparah oleh gejala gangguan pencernaan yang tumpang tindih, sehingga menyulitkan diagnosis dini.Untuk mengatasi tantangan ini, penelitian ini bertujuan untuk merancang dan membangun aplikasi sistem pakar berbasis web yang mampu mendeteksi gangguan pencernaan pada anak-anak. Sistem ini menerapkan metode Rule-Based Reasing dengan algoritma Forward Chaining, yang secara logis mencocokkan gejala yang dimasukkan pengguna dengan aturan dalam basis pengetahuan ini untuk menghasilkan diagnosis awal yang cepat. Aplikasi ini berfokus pada tiga indikator utama gangguan pencernaan: Diare, Konstipasi, dan GERD, untuk anak usia 5 bulan hingga 12 tahun. Dengan demikian, sistem pakar ini diharapkan dapat menjadi sarana interaktif yang dapat meningkatkan pemahaman orang tua, memudahkan dalam mendiagnosis gangguan pencernaan pada anak sejak dini, serta menjadi referensi bagi tenaga kesehatan dan pengembangan sistem informasi kesehatan
PENGARUH VIRAL MARKETING, BRAND AMBASSADOR, DAN KEPERCAYAAN KONSUMEN TERHADAP KEPUTUSAN PEMBELIAN NATASHA SKINCARE DI KOTA BATAM
Full text linkIncreasingly fierce business competition is driving companies to innovate in digital marketing strategies. One approach used is viral marketing, brand ambassadors, and building consumer trust to influence purchasing decisions. This study aims to determine whether influencer endorsements, customer reviews, and perceived quality influence purchasing decisions for Skintific products on TikTok Shop, both individually and collectively. A systematic and structured quantitative approach was used, located in Batu Aji District, Batam City. The population included all Natasha Skincare users in the area, and a sample of 100 people was determined using a purposive sampling method. Respondents were active users of Natasha Lightening Night Cream. Data were collected through a questionnaire and analyzed using multiple linear regression. Viral marketing has been proven effective in increasing purchase intention through the distribution of interactive content on social media. Brand ambassadors contribute to building a positive image and emotional attachment to the product. Consumer trust is formed from perceptions of quality, transparency, and previous experience, which strengthen the tendency to purchase. All three complement each other in forming stronger purchasing decisionsIncreasingly fierce business competition is driving companies to innovate in digital marketing strategies. One approach used is viral marketing, brand ambassadors, and building consumer trust to influence purchasing decisions. This study aims to determine whether influencer endorsements, customer reviews, and perceived quality influence purchasing decisions for Skintific products on TikTok Shop, both individually and collectively. A systematic and structured quantitative approach was used, located in Batu Aji District, Batam City. The population included all Natasha Skincare users in the area, and a sample of 100 people was determined using a purposive sampling method. Respondents were active users of Natasha Lightening Night Cream. Data were collected through a questionnaire and analyzed using multiple linear regression. Viral marketing has been proven effective in increasing purchase intention through the distribution of interactive content on social media. Brand ambassadors contribute to building a positive image and emotional attachment to the product. Consumer trust is formed from perceptions of quality, transparency, and previous experience, which strengthen the tendency to purchase. All three complement each other in forming stronger purchasing decisions