Jurnal Teknologi

Deteksi Cyberbullying Menggunakan BERT dan Bi-LSTM

Fidya Farasalsabila — 2024-05-04

Cyberbullying is a digital problem that is not a new phenomenon. This existed before the advent of social networks, and cyberbullying has a wide impact, including a person's mental and physiological conditions such as sadness, anxiety and depression. The main objective of this research is to develop an effective cyberbullying detection system using natural language processing techniques. The method used in this research includes the application of the BERT (Bi-Directional Encoder Representations from Transformers) and Bi-LSTM (Bi-Directional Long Short-Term Memory) models as a deep learning approach to analyze text and detect cyberbullying behavior. This approach allows the system to understand complex language contexts and capture patterns that traditional methods may find difficult to identify. Testing was carried out using a dataset that included various types of Indonesian language texts containing cyber bullying acts. The research results show that the combination of BERT and Bi-LSTM is able to provide superior detection performance with a high accuracy rate of 90% and the ability to identify variations of cyber bullying. This research makes a significant contribution to efforts to protect individuals from the negative impacts of cyber bullying through the development of a sophisticated and adaptive detection system.

Analisis Ergonomi Pengaruh Beban Kerja Terhadap Kelelahan Pada Pekerja Pembuatan Batu Bata Secara Manual

Muhammad Yusuf — 2024-06-22

Brick is a material that can be used as a building material to make houses, multi-storey buildings and so on. Making bricks manually cannot be separated from the risk of fatigue due to the workload carried out in the brick making process repeatedly. This research identifies and analyzes the effect of workload on work fatigue, determining the level of work fatigue on the workload of manual brick making workers. The population of this study was all 32 male workers, by determining Body Mass Index, determining cardiovascular load, determining workload and determining work fatigue. Data processing and analysis techniques were carried out using multiple linear regression tests. Based on the research results, it shows that there is a significant influence between workload and work fatigue in manual brick making workers, it is known that the value of p = 0.000 or smaller than 0.05 (p < 0.05), and the average level is large. work fatigue is 62.5 with the work fatigue category being moderate.

Analisis Pemilihan Supplier pada Toko Akbar Telur dengan Menggunakan Metode AHP

Siti Kumala Zahra — 2024-05-26

Eggs are an important source of animal protein as they contain high quality protein and have high biological value. Egg consumption in Indonesia continues to increase every year. This creates opportunities and competition in the egg market, including in Bogor City. In this context, the selection of the right supplier is key in meeting market demand and increasing the existence of sellers. This research was conducted at Shop Akbar Telur in Bogor Suryakencana Market. Data was collected through direct observation and interviews with the shop owner. The AHP method was used to process qualitative and quantitative data to prioritize suppliers based on price, quality and quantity criteria. The results of the analysis using AHP show that price is the highest criterion in selecting egg suppliers at Akbar Telur, followed by quality and quantity. Suppliers with the highest value for each criterion are SJT1 (price), SJT1 (quality), and SJT1 (quantity). The consistency level of criteria comparison is also acceptable. In selecting egg suppliers, the recommended overall priorities are SJT1, SJT2, and SB. The results of this study contribute to the understanding of supplier selection strategies that can improve store efficiency and performance.

Penyisihan Kandungan Perak dalam Limbah Cair Pencucian Film Melalui Proses Biofiltrasi dengan Tanaman Eceng Gondok

Siti Diyar Kholisoh — 2024-06-03

Wastewater produced from the film washing process in photo studios generally has very high contents of heavy metals, such as silver (Ag). This situation can cause various health and environmental problems if it is not treated before disposal. Therefore, this research was aimed to reduce the silver content in the wastewater through biological processes. This study was conducted using a biofiltration process using water hyacinth plants. Film washing waste was first acclimatized into concentration of 2.5% and the initial silver content was measured. Afterwards, water hyacinth of 600, 900, 1200, and 1500 g, respectively, was planted into the wastewater. Waste samples were subsequently taken and analyzed every 2 days. The highest removal efficiency of 99.94% was yielded for 1500 g of plant. In this condition, the silver content in wastewater met the environmental quality standards.

Pengaruh Suhu Evaporasi-Kristalisasi dan Persentase Gula Batu terhadap Warna Produk Serbuk Jahe Merah (Zingiber officinale)

Beauty Aprillia — 2024-06-06

Red ginger (Zingiber officinale) is recognized as a medicinal plant with multiple benefits. The medicinal benefits of red ginger necessitate cautious handling and treatment to avoid damaging the compounds present in ginger. Producing red ginger powder to make an instant red ginger drink is one of the most frequent methods of ginger processing. The aim of this study is to investigate the impact of raising the temperature during the evaporation-crystallization process (at 80 and 90 ^oC). Then, the proportion of crystallization agent is rock sugar (at 60 and 100%), which alters the color of red ginger extract, both in terms of research outcomes and commercial products. Each specimen is provided in the form of a solid stage (red ginger powder) and a liquid phase, whereby the red ginger powder is dissolved in hot water (approximately 50 ^oC). The color of both products was analyzed using a chromameter. The results indicate that the product samples' color changed with the color components L (dark-light), a (red-green), and b (yellow-blue). These changes were caused by an increase or decrease in the amount of dye in the red ginger extract. This caused carotenoids (red) and oleoresin (yellow). Additionally, they indicated a Maillard reaction due to the increase in temperature during the evaporation-crystallization process. The optimum outcomes of this research were observed when fresh raw materials were exposed to 90 °C and 100% cube sugar content. Notably, the color values exhibited no significant discrepancies from those of commercially available red ginger products.

Battle Royale Console Game Performance Analysis (Case Study: East Jakarta, Indonesia)

Noorfi Azizah Rahim — 2024-06-13

Battle Royale (BR) genre games are still played and have massive players around the globe. They are characterized by survival and exploration elements, featuring last-man-standing gameplay that motivates players to be the final contestant in the game. This BR can now be played on single platforms and cross-platform. Apex Legends is a free-to-play hero shooter where competitors from across the frontier join forces to battle for glory, fame, and fortune. Apex Legends can be played on PlayStation, Xbox, Steam, and Switch. This research aims to understand the data captured by the game Apex Legends, played on the PlayStation 5 console and displayed on a 4K television. This data was then analyzed to measure the correlation with game performance during gameplay. This research used a quantitative method, primary data, and descriptive measurements. From the findings, it can be concluded that the FPS for this game performed well since it was still near 60 FPS (59.81) with a standard deviation of 1.07. IO could not be analyzed for network performance since no official parameters for this IO were found on the official web during the research. However, Packet Loss and Choke performed well, with an average of 0 and a standard deviation of around 0. The ping value is enormous, with a significant standard deviation gap; this can happen because the server distance is longer to Tokyo from Jakarta than Singapore. Even though the latency was high, gamers still found the matches enjoyable.

Pelacakan Titik Daya Maksimum Algoritma Incremental Conductance pada Photovoltaic Dipasang Paralel dan Seri

Hendi Matalata — 2024-06-23

Photovoltaic (PV) is an environmentally friendly renewable energy source. Photovoltaic when connected directly to the load, the power delivered to the load is not optimal. Therefore it is important to get maximum power from the PV system to increase efficiency. This study presents an MPPT system using the Incremental Conductance (IC) algorithm to analyze PV systems installed in parallel and installed in series using 3 PVs. This IC algorithm works by monitoring the current gradient to determine the direction of voltage change and providing a change in derivative voltage so that it can determine the speed of delivery. From the results of the study, the average power using 3 PVs installed in series was 95.4% and using 3 PVs installed in parallel was 88.44%. The algorithm (IC) has a faster response in tracking the maximum power point when installed in parallel even though its efficiency is lower. However, current variations when there are changes in solar radiation tend to be constant as long as the temperature conditions are close to stable.

Literature Study on the Development of Neural Networks For Weather Forecasting

Fail Amir — 2024-06-24

Weather prediction has always been crucial for individuals to make informed decisions and protect themselves from potential hazards. Achieving accurate weather forecasts has historically been a significant challenge. Modern weather forecasting has evolved to integrate sophisticated computer models, data from atmospheric balloons and satellites, and insights from local observations. These methods have resulted in fairly precise predictions. Most forecasting models depend on complex mathematical formulas, but Artificial Neural Networks (ANN) offer a dynamic alternative, adapting their structure based on incoming data. This research aimed to thoroughly evaluate the effectiveness of ANNs in weather prediction. It explored the advantages of ANNs over traditional models, reviewed a range of methodologies, and documented the latest advancements in the field. The ultimate goal was to consolidate research findings to highlight the strides made in enhancing weather forecasting through ANNs.

Penggunaan Analytical Hierarchy Process dan Analytic Network Process dalam Menentukan Supplier Bahan Baku

Cyrilla Indri Parwati — 2024-06-25

Industry player, navigates the competitive convection business by strategically selecting raw material suppliers, with a focus on fabric. Employing the Analytical Hierarchy Process (AHP), a decision analysis method by Thomas L. Saaty, the company evaluates criteria such as price, quality, delivery, service, and location. In parallel, the Analytic Network Process (ANP), a more intricate method by Saaty, is utilized to account for reciprocal relationships within decision-making hierarchies. The company evaluates three alternative fabric suppliers: Kumala Textile, PT Kaha Tex, and PT Ada Tex. AHP calculations assign priority values – Kumala Textile at 0.381187, PT Kaha Tex at 0.470966, and PT Ada Tex at 0.073924. Concurrently, ANP prioritizes alternatives differently: Kumala Textile at 0.315383, PT Kaha Tex at 0.565728, and PT Ada Tex at 0.118889. Considering the results, PT Kaha Tex emerges as the most optimal raw material supplier with a dual-method endorsement. Kumala Textile secures the second recommendation, while PT Ada Tex takes the third spot. Selection criteria for fabric raw material suppliers encompass price, quality, delivery, service, and location. Grounded in prioritization values and criterion relationships, the comprehensive recommendation is in favor of PT Kaha Tex as the most optimal fabric raw material supplier.

Pemanfaatan Mikrotik RB942-2ND Menggunakan Metode Firewall Filtering Untuk Keamanan Jaringan Dengan Model Forensik

Rahmat Sulaiman — 2024-07-01

Nowadays as demand for fast and effective services increases, global technological and telecommunications developments also increase. Every village office, one of which is Belo Laut, must have an internet network to serve the community in accessing community data, therefore the development of technology and telecommunications is also very important for the government. To create a safe and effective internet network, this network security system uses Mikrotik with the Firewall Filtering method. The research problem is how to design and build network security in the Belo Laut village office area using Mikrotik to prevent the negative impacts of the internet, such as sites with adult content. Layer 7 protocol configuration and Filter Rules are included in blocking. Based on the keywords that have been determined, the test results show that Mikrotik has succeeded in blocking sites or websites that contain adult content.

Pemodelan Identifikasi Objek Kendaraan Bermotor Menggunakan Faster Region based Convolutional Neural Network (R-CNN) Berbasis Python

Rosalia Arum Kumalasanti — 2024-07-16

The vehicles are currently experiencing a surge in number and variation. This is evident from the kinds of vehicles that are passing through the highway area. The rise in the number of motorized vehicles will surely give a squeeze to the traffic density. The increase in the number of motor vehicles is one of the biggest factors in the impact of the congestion. The congestion can also cause damage to the highway. It's supposed to be the focus of the local government in dealing with the problem. Each road point has its own potential, so it is necessary to have a calculation in identifying the number of vehicles and the type of vehicles that are slipped on the road. Motor vehicle identification can be solved using the Faster Region based Convolutional Neural Network approach. Faster R-CNN is a deep learning architecture used to detect inside computers. Research will run at several highway points to take samples of video at a certain time, for identified the type of vehicle. Vehicle labelling will facilitate the calculation of the number of vehicles crossing the road in a given unit of time. The vehicle identification needs are used to see the density of the highway so that it can help the local government in making the right decision or solution to reduce the traffic density. The results of research such as quantitative data can be easily used to give the right picture and decision.

Porositas Serbuk Jahe Merah (Zingiber officinale Roscae var. Rubrum) terhadap Variasi Suhu dan Kadar Gula Pasir

Kristin Valentina Sinabutar — 2024-06-25

Ginger is one of Indonesia's most abundant and widespread cultivated plants and is widely used in people's daily lives as a raw material and additive in the manufacture of food and beverages. One type of ginger that is in great demand in the herbal beverage industry is red ginger. The bioactive compounds found in ginger include 6-gingerol, 6-paradol, 6-methylshogaol, 8-methylshogaol, 8-gingerol, 10-gingerol and 6-shogaol. Red ginger contains various bioactive compounds with pharmacological effects such as analgesic, anti-inflammatory, anti-cancer, antioxidant and cardiotonic effects. Scanning Electron Microscopy (SEM) is a microscopy technique used to produce high-resolution images of the surface of an electron sample. This study aimed to analyze the porosity of red ginger powder against variations in temperature and sugar content. The temperatures were 80 and 90°C with a sugar content of 60 and 100%. The results showed that the increase and decrease in the percentage of porosity values were caused by differences in temperature variations applied to the crystallization process. The rate of red ginger powder's porosity that approaches the fresh red ginger rate is red ginger powder at 80°C and 60% crystallization agent. The research results contributed to the characteristics of the materials used and the development of red ginger drink powder products.