PIKSEL : Penelitian Ilmu Komputer Sistem Embedded and Logic

Storyboard Design of Android-Based Learning Multimedia Integration Application Using Standard Process Tutorial Model

Santi Purwanti — Wed, 29 Mar 2023 01:43:02 +0000

Bina Pribadi Islam is a subject taught in Islamic elementary schools with the aim of improving students' spiritual well-being through guidance from teachers or mentors. Islamic elementary schools have an educational system designed by the Ministry of National Education and adapted to Islamic law. In the process of learning Bina Pribadi Islam, there is a need for a strategy to increase students' motivation and interest by using a multimedia application that includes elements such as text, audio, graphics, video, and animation tailored to the standard process tutorial model. To create an interactive multimedia application, multimedia design should be prepared in accordance with the tutorial model. Therefore, to produce a good application, a storyboard design needs to be created first that is tailored to the tutorial model to facilitate the application's development. Based on the storyboard design created according to the tutorial model, three main menus are produced as the characteristic features of the tutorial model, namely introduction, tutorial material, and exercises. Each menu produces several sub-menus tailored to the material in the Bina Pribadi Islam subject.

Pi Hole on SOE Computer Network using Raspberry Pi 3 Model B+ to Optimize Bandwidth Management and Improve Employee Performance

Rahmat Novrianda Dasmen, Darwin Darwin, Irham Irham, Bima Riansyah — Wed, 29 Mar 2023 00:00:00 +0000

State-Owned Enterprises (BUMN) are government agencies that currently have computer networks and are supported by internet services, which are always used in the work process of BUMN employees. Along with the development of the Internet of Things (IoT) devices, good internet service is very much needed by BUMN as it will affect the performance of BUMN employees. However, the problem that often occurs in internet services in BUMN is slow internet speed, which hinders the work process of BUMN and decreases the performance of BUMN employees. It should be noted that good internet service depends on the bandwidth used, so it is very important to have good bandwidth management applied to the BUMN computer network. So far, BUMN have done several things related to bandwidth management, ranging from access restrictions, even distribution of bandwidth, and the implementation of an authentication system to connect to the BUMN internet network. However, some of these things are still not enough to guarantee no wasteful use of bandwidth, which occurs because when operating some work-supporting websites, various types of digital ads will automatically appear, consuming a significant amount of bandwidth. Therefore, in this Action Research study, a Pi Hole to control digital ads was implemented that can automatically appear and use large amounts of bandwidth. By using Raspberry Pi 3 Model B+, the development of Pi Hole will limit digital ads from appearing when accessing websites or other IoT devices used by BUMN so that bandwidth usage will be more optimal in BUMN work processes. This will also improve the performance of BUMN employees, where with maximum bandwidth, the work process of BUMN can be accelerated.

The Influence of Youtube Ads on Purchase Intention

Wed, 29 Mar 2023 02:07:40 +0000

The high number of social media users on YouTube has made many companies interested in advertising on the platform. However, many YouTube users tend to skip the ads that are displayed. The main objective of this research is to determine the factors within YouTube ads that have an influence on Purchase Intention. This study uses a quantitative research method, with variables including Informativeness, Entertainment, Customization, Irritation, Advertising Value, Flow Experience, Brand Awareness, and Purchase Intention. The results of this research show that Informativeness has a significant influence on Advertising Value but not on Flow Experience. Meanwhile, Entertainment and Customization have a significant influence on both Advertising Value and Flow Experience, and Advertising Value, Brand Awareness, and Flow Experience have a significant influence on Purchase Intention. As for Irritation, this variable does not have a significant influence on Advertising Value and Flow Experience.

Optimization of Random Forest Prediction for Industrial Energy Consumption Using Genetic Algorithms

Wed, 29 Mar 2023 02:37:48 +0000

Abstract

Saving electrical energy consumption in industries is crucial; hence, the prediction of industrial energy consumption needs to be performed. The random forest method can be applied to steel industry data to predict energy consumption. The purpose of this prediction is to increase energy savings in industries and optimize the performance of the random forest method. The results of the random forest show that the algorithm can predict energy consumption in industries effectively; however, it needs further optimization to achieve better predictions. Therefore, the genetic algorithm method will be used to optimize the previous method. The optimization results indicate that it is successfully conducted in terms of accuracy and kappa level. This optimization is beneficial to society, especially industrial companies.

Cluster Analysis Using Principal Component Analysis Method and K-Means to Find Out the Compliance Group of Property Tax

Wed, 29 Mar 2023 03:00:54 +0000

Abstract

The village of Kendal has experienced a decline in local income due to the high rate of property tax arrears, with 226 taxpayers (19% of residents) known to have outstanding payments. Additionally, with 1,159 separate residents residing in 10 block areas with varying tax amounts, it has become increasingly difficult for the Village Apparatus to profile taxpayers based on their characteristics. To overcome these problems, a data analysis model based on Machine Learning technology will be developed using the Principal Component Analysis (PCA) Method combined with the K-Means method. The objective of this study is to create a cluster analysis model that can accurately map the characteristics of taxpayers, making it easier for the Village Apparatus to identify and assist residents who need to pay their property tax. This proposed solution will also simplify the reporting process to the central government regarding the estimated regional revenue sourced from property tax.

EfficientNetV2M for Image Classification of Tomato Leaf Deseases

Wed, 29 Mar 2023 03:29:35 +0000

Favorable climatic conditions make tomato plants (Solanum Lycopersicon) a widely cultivated horticultural crop in Indonesia. However, the increase in tomato production is often accompanied by a decrease in both the quantity and quality of the plants, which can be caused by a variety of factors such as bacteria, fungi, viruses, and insects like Late Blight and Two-Spotted Spider Mite diseases that attack the tomato leaves. To help farmers identify leaf diseases that have similar characteristics, this study employs image processing with the Convolutional Neural Network (CNN) algorithm and transfer learning models. Specifically, the study uses the EfficientNetV2M transfer learning architecture which has superior parameter efficiency and training speed compared to other transfer learning models. Additionally, this study conducts four experimental scenarios on preprocessing, including green channel + CLAHE, green channel + Gaussian Blur, CLAHE without green channel, and Gaussian Blur without green channel. The dataset used in this study includes 5,176 images with three labels: Tomato Healthy, Tomato Late Blight, and Tomato Two-Spotted Spider Mite. These images were used to train and produce models, which were then tested using a different dataset from the trained dataset. The testing dataset included 30 image samples divided into three labels. Based on the test results of the four models with different scenarios, the best model was found to be the one with the green channel preprocessing scenario + CLAHE, which was able to precisely predict all 30 image samples with high accuracy.

Identification of Website-Based Product Sales Frequency Patterns using Apriori Algorithms and Eclat Algorithms at Rio Food in Bekasi

Salwa Nabiila Pramuhesti, Herlawati Herlawati, Tyastuti Sri Lestari — Wed, 29 Mar 2023 04:18:12 +0000

Sales reports that are not managed automatically may hinder businesses from accurately determining their progress in the short or long term. With increasing community needs for a product, business owners have an opportunity to market their products to a larger audience. The abundance of data highlights the need for information to produce patterns that can be used as a reference for making decisions in buying products on the website. Data mining algorithms can provide support for analysis, which can help avoid inaccurate business progress reports. In this study, the Apriori and Eclat algorithms were applied to analyze frequent itemsets in association rule mining. The dataset used in this study consists of 20 transaction data from frozen food sales. The results showed that the combination of Nugget and Chicken Sausage itemsets were the most frequent, with higher support, confidence, and lift ratio values than the others. These results can be used as product recommendations that are most in demand by customers.

The Weighted Product Method and the Multi-Objective Optimization on the Basis of Ratio Analysis Method for Determining the Best Customer

Mugiarso Mugiarso, Rasim Rasim — Fri, 31 Mar 2023 00:00:00 +0000

The objective of this study is to compare the effectiveness of the Weighted Product (WP) and Multi-Objective Optimization on the Basis of Ratio Analysis (MOORA) methods in determining the best customers. Onesnet, the case study service provider, provides discounts and rewards to eligible customers to support this objective. The problem addressed in this study is how to determine the most relevant method for selecting eligible customers for bonuses. To achieve this, sensitivity testing was conducted by altering the weights of each criterion in both methods and observing the percentage changes of the results. The Weighted Product method multiplies the rating of each connected attribute, which is raised to the appropriate attribute weight, to decide. Data for this study was collected through interviews and observations at Onesnet and processed using the Rank Order Centroid (ROC) method for weighting, and the WP and MOORA methods for evaluating and selecting a decision. The WP and MOORA methods produced different total values and rankings, but the modeling with either method can be used equally for selecting the best customers. While there was a 60% similarity in data between the two methods, the WP method is recommended over MOORA, as it prioritizes customers with high loyalty criteria as the best customers.

Factors Influencing Students' Intention to use Online Tutoring Applications in Jakarta

RA Dyah Wahyu Sukmaningsih, Adam Kurniawan, Ronald Ronald — Fri, 31 Mar 2023 00:00:00 +0000

As the COVID-19 pandemic has disrupted traditional learning methods, many students have turned to online tutoring as a supplementary source of education. This study aims to identify the factors that influence students' intention to use online tutoring applications. Data was collected from 401 student respondents in Jakarta through a questionnaire, and analyzed using smart PLS. The results show that perceived brand orientation, interactive course features, course quality, perceived usefulness, perceived ease of use, and trust all have a significant impact on students' intention to use online tutoring applications. These findings have implications for the design and promotion of online tutoring applications, as well as for policymakers and educators seeking to support student learning in the era of COVID-19.

Decision Support System Design for Informatics Student Final Projects Using C4.5 Algorithm

Rafika Sari, Hasan Fatoni, Khairunnisa Fadhilla Ramdhania — Fri, 31 Mar 2023 00:00:00 +0000

Academic consultation activities between students and academic supervisors are necessary to help students carry out academic activities. Based on the transcript of grades obtained, many students do not choose the appropriate final project/thesis specialization fields based on their academic abilities, resulting in a lot of inconsistencies between the course grades and the final project specialization fields. The purpose of this research is to minimize the subjectivity aspect of students in choosing their final project academic supervisors and minimize the inconsistencies between the course grades and the final project specialization fields. The method used in this research is classification data mining using the Decision Tree and C4.5 Algorithm methods, with the attributes involved being courses, course grades, and specialization courses. The C4.5 Decision Tree algorithm is used to transform data (tables) into a tree model and then convert the tree model into rules. The implementation of the C4.5 Decision Tree algorithm in the specialization field decision support system has been successfully carried out, with an accuracy rate of 70% from the total calculation data. The data used in this research is a sample data from several senior students in the Informatics program at Ubhara-Jaya. The results of the research decision support system can be used as a good recommendation for the Informatics program and senior students to direct their final project research. It is expected that further research will use more sample data so that the accuracy rate will be better and can be implemented in website or mobile-based applications.

User Experience Evaluation on Production Performance Monitoring System Using Honeycomb Method

Moh. Sofyan Sauri, Arie Hidaya Putra, Emny Harna Yossy — Fri, 31 Mar 2023 00:00:00 +0000

In the era of digitalization in industries, companies are implementing the Digital Factory Operating System in various aspects, including monitoring production performance on the sachet line. Previously, operators manually carried out the production performance monitoring process using paper forms, and admins re-entered the data into the system. Although the monitoring process has begun to be carried out digitally using available tablets, its usage still needs improvement. Therefore, it is necessary to evaluate the user experience of the Production Performance Monitoring Information System to determine user satisfaction. The Honeycomb method is used in this study, which assesses seven aspects, namely accessible, credible, desirable, findable, usable, useful, and valuable. The results show the average final score of each aspect assessed, and the highest score of 66.45% is for the accessible and useful aspects. The study shows that the user experience evaluation of the Production Performance Monitoring Information System using the honeycomb method is generally good, but improvements are still needed in some aspects, such as accessibility and desirability.

Usability Analysis on Health Tracking Application using User Experience Questionnaire in Jakarta Area

Richard Richard, Aditya Kusumadwiputra, Adela Zahwa Firdaus Suherman — Fri, 31 Mar 2023 00:00:00 +0000

The study aims to investigate user satisfaction with the Health Tracking application in Jakarta, which is a vital tool used by the government to monitor the health of Indonesians. The research will employ a User Experience Questionnaire to gather feedback from a minimum of 400 respondents who are users of the Health Tracking application selected through random sampling. The analysis of the survey results will evaluate the User Interface (UI) and User Experience (UX) of the application and provide valuable insights that can be used to enhance the app's future development. The UEQ Data Analysis Tool will be utilized to analyze data. Based on the findings, it can be concluded that users are highly satisfied with the Health Tracking application's UI/UX. However, improvements can be made to enhance the perspicuity aspect of the app, along with maintaining or improving other factors. The results of this study can serve as a benchmark for the future development of the Health Tracking application.

Identifying Factors Affecting the Relationship between Department and Graduation Level of Informatics Engineering Students using Apriori Algorithm: A Case Study at Pamulang University

Thoyyibah T — Fri, 31 Mar 2023 00:00:00 +0000

To cultivate the next generation of leaders, it is essential for teenagers to receive a high level of education. Typically, this education is acquired through attending lectures that produce a high GPA, which is considered a valuable achievement for students. The level of graduation achieved within the appropriate timeframe can also impact campus accreditation, especially for engineering students, particularly those pursuing informatics engineering. To improve graduation rates, it is necessary to use data mining to identify patterns and trends among graduating students. The a priori algorithm was used in this study to analyze school majors, the length of study, and student graduation rates. Through this algorithm, it was possible to identify one or more rules that can be used as benchmarks for predicting graduation rates. Based on the results and discussions of 30 students, the most effective rule for predicting graduation is a combination of the student's previous school major, a study period of 4 years or less, a GPA of 2.51-3.00, and passing all courses on time. Using the a priori algorithm, the rule was found to have a confidence value of 16 and a support value of 71.4%. This indicates that the rule is a reliable predictor of student graduation rates.

Sentiment Analysis of Sentence-Level using Dependency Embedding and Pre-trained BERT Model

Fri, 31 Mar 2023 00:00:00 +0000

Sentiment analysis is a valuable field of research in NLP with many applications. Dependency tree is one of the language features that can be utilized in this field. Dependency embedding, as one of the semantic representations of a sentence, has shown to provide more significant results compared to other embeddings, which makes it a potential way to improve the performance of sentiment analysis tasks. This study aimed to investigate the effect of dependency embedding on sentence-level sentiment analysis through experimental research. The study replaced the Vocabulary Graph embedding in the VGCN-BERT sentiment classification system architecture with several dependency embedding representations, including word vector, context vector, average of word and context vectors, weighting on word and context vectors, and merging of word and context vectors. The experiments were conducted on two datasets, SST-2 and CoLA, with more than 19 thousand labeled sentiment sentences. The results indicated that dependency embedding can enhance the performance of sentiment analysis at the sentence level.

Fuzzy Use Case Points as a Basis for Effort Estimation

Rakhmi Khalida, Tubagus Maulana Kusuma, Khairunnisa Fadhilla Ramdhania — Fri, 31 Mar 2023 00:00:00 +0000

Many software development projects encounter problems related to over- or under-estimation of effort. Accurate effort estimation is crucial for successful project management, but it can be challenging when resources are limited, and little is known about the project. The commonly used method for effort estimation is Use Case Points (UCP), which is mainly used for application-based objects and takes use cases as input. However, UCP has weaknesses, particularly in the high variation of weight factor values for Unadjusted Use Case Weight (UUCW). To address this problem, Fuzzy Use Case Points (FUCP), which is a combination of fuzzy logic and use case points, can be used. By applying fuzzy logic to the UUCW category, FUCP derives new weight factor values for UUCW. The implementation of FUCP to calculate effort estimation in ten government-based projects in this research has shown that FUCP yields the closest value to the actual effort required. It has also been demonstrated that FUCP outperforms UCP in terms of accuracy, with an improvement of 6.51%.

Learning Vector Quantization, Hebbian Learning, and Self-Organizing Map for Classification

Herlawati Herlawati — Fri, 31 Mar 2023 00:00:00 +0000

Deep Learning has been rapidly developed. Almost all proposed methods already have very high accuracy. Most of these methods still use techniques from the past with some modifications to adapt to existing modules. Sometimes it is necessary to understand past methods to produce new methods. Therefore, this research examines past models that have the potential to improve the performance of existing deep learning models. The methods to be examined include Learning Vector Quantization (LVQ), Hebbian learning, and Self-Organizing Map (SOM). The iris dataset available on Scikit-learn (SKlearn) is used here for testing in cases of supervised learning and unsupervised learning (especially SOM). The results show that LVQ has a good accuracy of 93%, while Hebbian learning has an accuracy of 56%. SOM fluctuates between 88% and 93%. Although the accuracy of SOM does not exceed LVQ, this model does not require labels in its training process.