cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
-
Contact Email
-
Phone
-
Journal Mail Official
sji@mail.unnes.ac.id
Editorial Address
-
Location
Kota semarang,
Jawa tengah
INDONESIA
Scientific Journal of Informatics
Published by Universitas Negeri Semarang
ISSN : 24077658     EISSN : 24600040     DOI : -
Core Subject : Science,
Computer Science & IT
Scientific Journal of Informatics published by the Department of Computer Science, Semarang State University, a scientific journal of Information Systems and Information Technology which includes scholarly writings on pure research and applied research in the field of information systems and information technology as well as a review-general review of the development of the theory, methods, and related applied sciences.
Arjuna Subject : -
Articles 15 Documents
 1   2 
Search results for , issue "Vol 10, No 3 (2023): August 2023" : 15 Documents clear
Coastal Sentiment Review Using Naïve Bayes with Feature Selection Genetic Algorithm Somantri, Oman; Maharrani, Ratih Hafsarah; Purwaningrum, Santi
Scientific Journal of Informatics Vol 10, No 3 (2023): August 2023
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v10i3.43988

Abstract

Purpose: The tourism potential in the maritime sector can be Indonesia's mainstay at this time, especially in enjoying the charm of the natural beauty of the coast as people know Indonesia is an archipelagic country. The purpose of this study is to find the best model by applying the feature selection genetic algorithm (GA) and Information Gain (IG) to get the best Naïve Bayes (NB) model and the best features to produce the best level of sentiment classification accuracy.Methods: The stages of the research were carried out by going through the process of searching, pre-processing, analyzing research data using the Naïve Bayes model and optimizing genetic algorithms, validating data, and model evaluation.Result: The experimental results show that the best model is naïve Bayes based on information gain and the genetic algorithm yields an accuracy rate of 86.34%.Novelty: The main contribution to this research is proposing a new model of the best NB optimization model by applying an optimization algorithm in the search for feature selection to increase sentiment classification accuracy.
Alphabet Classification of Sign System Using Convolutional Neural Network with Contrast Limited Adaptive Histogram Equalization and Canny Edge Detection Raharjo, Ahmad Solikhin Gayuh; Sugiharti, Endang
Scientific Journal of Informatics Vol 10, No 3 (2023): August 2023
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v10i3.44137

Abstract

Purpose: There are deaf people who have problems in communicating orally because they do not have the ability to speak and hear. The sign system is used as a solution to this problem, but not everyone understands the use and meaning of the sign system, even in terms of the alphabet. Therefore, it is necessary to classify a sign system in the form of American Sign Language (ASL) using Artificial Intelligence technology to get good results.Methods: This research focuses on improving the accuracy of ASL alphabet classification using the VGG-19 and ResNet50 architecture of the Convolutional Neural Network (CNN) method combined with Contrast Limited Adaptive Histogram Equalization (CLAHE) to improve the detail quality of images and Canny Edge Detection to produce images that focus on the objects in it. The focused result is the accuracy value. This study uses the ASL alphabet dataset from Kaggle.Result: Based on the test results, there are three best accuracy results. The first is using the ResNet50 architecture, CLAHE, and an image size of 128 x 128 pixels with an accuracy of 99.9%, followed by the ResNet50 architecture, CLAHE + Canny Edge Detection, and an image size of 128 x 128 pixels with an accuracy of 99.82 %, and in third place are the VGG-19 architecture, CLAHE, and an image size of 128 x 128 pixels with an accuracy of 98.93%.Novelty: The novelty of this study is the increase in the accuracy value of ASL image classification from previous studies.
Fuzzy Smart Reward for Serious Game Activity Design Haryanto, Hanny; Rosyidah, Umi; Kardianawati, Acun; Astuti, Erna Zuni; Dolphina, Erlin; Haryanto, Ronny
Scientific Journal of Informatics Vol 10, No 3 (2023): August 2023
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v10i3.44051

Abstract

Purpose:  Serious game has been widely considered to be a potential learning tool, due to its main advantage to provide a fun experience in learning. The experience is supported mainly by in-game activities, where feedback is given in the form of rewards. However, rewards often don't work well due to various factors, for example, rewards are always the same, so they are monotonous. We use Appreciative Learning as underlying concept for activity design and fuzzy logic to create the reward behavior, called Fuzzy Smart Reward.Methods: We use Appreciative Learning as underlying concept for activity design and fuzzy logic to create the reward behavior. Appreciative Learning activities consists of Discovery, Dream, Design and Destiny. We propose fuzzy-based smart reward for those activities. The smart reward takes player achievement in each activity as input for the fuzzy inference system and give the dynamic reward as output.Result: A game prototype is developed as a test subject. The result shows that the smart reward could dynamically adjust the reward based on game conditions and player performance. Test conducted using Game Experience Questionnaire get the score 3.3 out of 4.Novelty:  There aren't many studies on dynamic rewards in structured reward systems; the majority of studies remove dynamic rewards from reward systems. In our research, a "smart reward" is a dynamic reward in a structured reward system that is created using artificial intelligence and is based on activities for appreciative learning. The use of Fuzzy Logic for structured reward behavior is also very rare. 
Capital Optical Character Recognition Using Neural Network Based on Gaussian Filter Astuti, Erna Zuni; Sari, Christy Atika; Syabilla, Mutiara; Sutrisno, Hendra; Rachmawanto, Eko Hari; Doheir, Mohamed
Scientific Journal of Informatics Vol 10, No 3 (2023): August 2023
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v10i3.43438

Abstract

Purpose: As digital technology advances, society needs to convert physical text into digital text. There are now many methods available for doing this. One of them is OCR (Optical Character Recognition), which can scan images [1]–[4] containing writing and turn them into digital text, making it easier to copy written text from an image. Text recognition in images is complex due to variations in text size, color, font, orientation, background, and lighting conditions.Methods: The technique of text recognition or optical character recognition (OCR) in images can be done using several methods, one of which is a neural network or artificial neural network. The artificial neural network method can help a computer make intelligent decisions with limited human assistance. Intelligent decisions can be made because the neural network can learn and model the relationship between nonlinear and complex input and output data. In this research, the scaled conjugated gradient is applied for optimization. SCG is very effective in finding the minimum value of a complex function, but it takes longer than some other optimization algorithms.Result/Findings: The dataset used is an image with a size of 28 x 28 which is changed in dimension to 784 x 1. This research uses 4000 epochs and obtained the best validation result at epoch 3506 with a value of 0.0087446. Results: From the statistical test results, the effect of perceived usefulness on ease of use has the highest level of influence, obtaining a test value of 3.6. Furthermore, the effect of the attitude towards using on the behavioral intention to use has the lowest level of influence, which obtained a test value of 1.2.Novelty:  In this article, Gaussian filter is used as feature extraction to improve yield. Character detection results using a Gaussian filter are known to be almost 10% higher than those using only a neural network. The result with the Neural Network alone is 82.2%, while the Neural Network-Gaussian Filter produces 92.1%.
Usability Analysis of Digital-Based Agricultural Product Marketing Platform at Farmers Level in Region V, Bogor Regency Budiastuti, Evrina; Ritchi, Hamzah; Deliana, Yosini
Scientific Journal of Informatics Vol 10, No 3 (2023): August 2023
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v10i3.44605

Abstract

Purpose:  This study examines the usability of a digital platform called Kiosagri.com. This study needs to carry out as one of the solutions to overcome the limited accessibility of agricultural product promotion through the use of a digital platform that is expected to have usability. Therefore, this study aims to analyze the usability of digital-based agricultural product marketing platforms at the farmer level in Region V of Bogor Regency.Methods:  This research employed a mixed method with an explanatory sequential design. The intended audience for the census of respondents was farmers who had utilized the Kiosagri.com platform. Data analysis was done on 32 respondents during the quantitative phase using descriptive statistics. Ten respondents were chosen purposefully for descriptive qualitative analysis during the qualitative phase.Results:  According to the results, the Kiosagri platform is very user-friendly. It is achieved as a result of several beneficial aspects connected to the benefits perceived by users, such as ease of use, suitability of design, and features offered by Kiosagri.Novelty:  In this study, a different way to measure the usability of digital platforms for agricultural products marketing is applied, namely by using the explanatory sequential design by conducting usability testing using descriptive statistics through the use of questionnaires and explaining the results further using descriptive qualitative through in-depth interviews which both methods are suitable for testing user experience and usability in the agricultural sector. The results of this study will be useful for the government in implementing digital marketing of agricultural products.
Performance Improvement of Fake News Detection Models Using Long Short-Term Memory Hyperparameter Optimization Lindawati, Lindawati; Ramadhan, Muhammad Fadli; Soim, Sopian; Novianda, Nabila Rizqi
Scientific Journal of Informatics Vol 10, No 3 (2023): August 2023
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v%vi%i.45420

Abstract

Purpose: The proposed model was developed based on prior research that distinguished between fake and real news using a deep learning-based methodology and an LSTM neural network, with a model accuracy of 99.88%. This study uses hyperparameter tuning techniques on a Long Short-Term Long Memory (LSTM) neural network architecture to improve the accuracy of a fake news detection model.Methods: To improve the accuracy of the fake news detection model and optimize the model from previous research, this study uses the hyperparameter tuning technique on models with Long Short-Term Memory (LSTM) neural network architecture. For this technique, three different types of experiments, hyperparameter tuning on the LSTM layer, Dense layer, and Optimizer, were conducted to obtain the best hyperparameters in each layer of the model architecture and the model parameters proposed. The fake and real news dataset, which has also been used in earlier studies, was used in this study.Results: The proposed model could detect fake news with a high accuracy of 99.97%, surpassing the previous research models with an accuracy of 99.88%.Novelty: The novelty of this study was the hyperparameter tuning technique on different layers of the LSTM neural network to optimize the fake news detection model. The research aims to improve upon previous approaches and increase the accuracy of the model. 
A Comparative Analysis on the Evaluation of KNN and SVM Algorithms in the Classification of Diabetes Limas, Agus Fahmi; Rosnelly, Rika; Hartono, Hartono; Nursie, Aly
Scientific Journal of Informatics Vol 10, No 3 (2023): August 2023
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v10i3.44269

Abstract

Purpose: Diabetes has received a great deal of attention in medical research because of its profound effect on human health. Many factors cause this disease in the human body. Can be from food or drink that is often consumed by the human body. Diabetes cannot be cured and can only be controlled.Methods: In this study, using 2 data mining techniques namely Support Vector Machine and K-Nearest Neighbor were applied to predict diabetes. In this study, 768 diabetes data were used as trial data, consisting of training data that had been pre-processed data and 400 data cleaning data, 278 data testing data, and 50 diabetes data samples used as samples in the calculation.Result: The performance of each algorithm is analyzed differently, the results of each best algorithm will be analyzed to determine which algorithm can provide better results for predicting diabetes. The results obtained in this study get a value of 0 where the predicted value of the target class for new data is the negative class (Suffer).Novelty: This study compares the SVM and K-NN methods for diabetes classification. So, successfully implemented for data on the classification target
Comparison of ARIMA and GRU Models for High-Frequency Time Series Forecasting. Ridwan, Mochamad; Sadik, Kusman; Afendi, Farit Mochamad
Scientific Journal of Informatics Vol 10, No 3 (2023): August 2023
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v10i3.45965

Abstract

Purpose: The purpose of this research is to assess the efficacy of ARIMA and GRU models in forecasting high-frequency stock price data, specifically minute-level stock data from HIMBARA banks. In time series analysis, time series data exhibit interesting interdependence among observations. Despite its popularity in time series forecasting, the ARIMA model has limitations in capturing complicated nonlinear patterns. Forecasting high-frequency data is becoming more popular as technology advances and more high-frequency data becomes available.Methods: In this study, we compare the ARIMA and GRU models in forecasting minute-level stock prices of HIMBARA banks. The data used consists of 62,921 minute-level stock data points for each bank in the HIMBARA group, collected in the year 2022. The GRU model was chosen because it is capable of capturing complex nonlinear patterns in time series data. Each method's predicting performance is assessed using the Mean Absolute Percentage Error (MAPE) statistic.Results: In terms of forecasting accuracy, the GRU model outperforms the ARIMA model. The GRU model achieves a MAPE of 0.77% for BMRI stock, while the ARIMA model achieves a MAPE of 4.09%. The GRU model predicts a MAPE of 0.34% for BBRI stock, while the ARIMA model predicts a MAPE of 3.02%. For BBNI stock, the GRU model obtains a MAPE of 0.63%, while the ARIMA model achieves a MAPE of 1.52%. The GRU model achieves a MAPE of 0.58% for BBTN stock, while the ARIMA model achieves a MAPE of 6.2%.Novelty: In terms of minute-level time series data modeling, research in Indonesia has been limited. This study adds a new perspective to the discussion by comparing two modeling approaches: the traditional ARIMA model and the sophisticated deep learning GRU model, both of which are applied to high-frequency data. Beyond the present scope, there are several promising future directions to pursue, such as anticipating intraday stock fluctuations. This unexplored zone not only contributes to the field of financial modeling but also has the ability to uncover intricate patterns in minute-level data, an area that has not been extensively studied in the Indonesian context.
Clothing Sales Prediction Information System Using Web-Based Double Exponential Smoothing Method Sitorus, Apriyanti Anggraini; Ikhwan, Ali; H. Aly, Moustafa
Scientific Journal of Informatics Vol 10, No 3 (2023): August 2023
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v10i3.44919

Abstract

Purpose: The purpose of this research is to determine the smallest error value so that the resulting prediction data is more accurate. This prediction data is used to help Raja Fashion Medan in processing goods data and help predict the amount of goods that must be provided to meet customer needs in the next period.Methods: This research uses the Double Exponential Smoothing method because it is used on data that is more stable and has a trend pattern. To test the accuracy of the prediction results with the Double Exponential Smoothing method, the Root Mean Square Error (RMSE) and Mean Absolute Percent Error (MAPE) data testing methods are used by finding the smallest error value.Result: This test is carried out by determining the smallest error value on 118 data types of goods with error results, namely the average Root Mean Square Error (RMSE) of 26.5, Mean Absolute Deviation (MAD) 1.2, Mean Squared Error (MSE) 37.8 and Mean Absolute Percent Error (MAPE) of 10%, it can be concluded that the accuracy of theprediction is very good.Novelty: Testing on prediction results uses 4 methods to determine more accurate results, namely with Root Mean Square Error (RMSE), Mean Absolute Deviation (MAD), Mean Squared Error (MSE), and Mean Absolute Percent Error (MAPE) which are used to find values smallest error.
Public Satisfaction on Online Service Development: (Case Study: Bantulpedia Application) Sukarno, Mohamad; Pribadi, Ulung
Scientific Journal of Informatics Vol 10, No 3 (2023): August 2023
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v10i3.45985

Abstract

Purpose: This study analyzes public satisfaction with Bantulpedia Application users in Bantul Regency. The measurement was carried out using variables from the Online Service Index (OSI) framework consisting of institutional framework, service provision, content provision, technology, and e-participation.Methods: The research method used is quantitative, with the primary data source being a questionnaire totaling 100 respondents. SmartPLS 3 software is used in conducting data analysis for this research.Results: The research results show that variable institutional framework has a positive and significant influence on the satisfaction of Bantulpedia Application users. Meanwhile, the variables service provision, content provision, technology, and e-participation did not positively and significantly influence the satisfaction of Bantulpedia application users.Novelty: This study is unique by using OSI parameters by correlating them with application-based public satisfaction assessments (online services), which the majority of this theory is only used to measure e-government development and various aspects of it. This research contributes to providing a new perspective on using OSI theory in the realm of online services (applications) not only as a predictive measurement but as a significant research object (a country). In addition, the empirical contribution is to conduct a public satisfaction test on using the Bantulpedia Application. So that later it can also be adopted and provide space for other local governments to adopt OSI as an indicator of this online service.

Page 1 of 2 | Total Record : 15

 1   2 

Filter by Year

2023 2023


Reset
Filter By Issues
All Issue Vol 11, No 1 (2024): February 2024 Vol 10, No 4 (2023): November 2023 Vol 10, No 3 (2023): August 2023 Vol 10, No 2 (2023): May 2023 Vol 10, No 1 (2023): February 2023 Vol 9, No 2 (2022): November 2022 Vol 9, No 1 (2022): May 2022 Vol 8, No 2 (2021): November 2021 Vol 8, No 1 (2021): May 2021 Vol 7, No 2 (2020): November 2020 Vol 7, No 1 (2020): May 2020 Vol 6, No 2 (2019): November 2019 Vol 6, No 1 (2019): May 2019 Vol 6, No 1 (2019): Mei 2019 Vol 5, No 2 (2018): November 2018 Vol 5, No 2 (2018): November 2018 Vol 5, No 1 (2018): May 2018 Vol 5, No 1 (2018): May 2018 Vol 4, No 2 (2017): November 2017 Vol 4, No 2 (2017): November 2017 Vol 4, No 1 (2017): May 2017 Vol 4, No 1 (2017): May 2017 Vol 3, No 2 (2016): November 2016 Vol 3, No 2 (2016): November 2016 Vol 3, No 1 (2016): May 2016 Vol 3, No 1 (2016): May 2016 Vol 2, No 2 (2015): November 2015 Vol 2, No 2 (2015): November 2015 Vol 2, No 1 (2015): May 2015 Vol 2, No 1 (2015): May 2015 Vol 1, No 2 (2014): November 2014 Vol 1, No 2 (2014): November 2014 Vol 1, No 1 (2014): May 2014 Vol 1, No 1 (2014): May 2014 More Issue