Garuda - Garba Rujukan Digital

Analisis Prediksi Kata Kunci Situs Web MonsterMAC dengan Metode Long Short-Term Memory (LSTM) Hanif Assalmi, Fityan; Syaifullah Jauharis Saputra, Wahyu; Muhaimin, Amri
Jurnal Teknologi Terpadu Vol 10 No 1 (2024): Juli, 2024
Publisher : LPPM STT Terpadu Nurul Fikri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54914/jtt.v10i1.1187

Amid increasingly fierce competition in the digital realm, many companies are striving to increase the number of visitors to their websites. One such competing company is MonsterMAC, a startup. This research aims to provide early warnings and analyze relevant keywords on the MonsterMAC website using the Long Short-Term Memory (LSTM) method. Visitor data from Google Analytics and keyword data from Google Trends for the period July 22, 2022, to July 15, 2023, have been collected and processed through several stages, such as preprocessing, model design, LSTM training, and testing, as well as visualization and interpretation of results. The modeling results show satisfactory performance, with MAE Train Real User = 0.0615, Vending Machine = 0.0218, IoT = 0.0284, Machine Learning = 0.0365, Digital Business = 0.0186, Business Intelligence = 0.0296. Furthermore, this research indicates that the number of visitors is predicted to increase but will also experience a sharp decline in the coming days. The use of the keyword "IoT" shows a significant increasing trend. Implementing the keyword "IoT" in SEO strategies has increased the number of visitors over the next seven days from 250 to 350. This research guides website owners in optimizing their content and SEO strategies to increase their visibility and competitiveness in a highly competitive digital environment. This research also emphasizes the importance of the LSTM method in keyword analysis and prediction to create more targeted SEO strategies.

Prediction of The Islamic Stock Price Index and Risk of Loss Using The Long Short-Term Memory (LSTM) and Value At Risk (VaR) Taufik, Ikbar Athallah; Trimono, Trimono; Muhaimin, Amri
IJDASEA (International Journal of Data Science, Engineering, and Analytics) Vol. 4 No. 01 (2024): International Journal of Data Science, Engineering, and Analytics Vol 4, No 1,
Publisher : Universitas Pembangunan Nasional Veteran Jawa Timur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33005/ijdasea.v4i01.16

Investment aims to increase the value of capital or earn additional income through asset growth, dividends or profits. One investment instrument that is in demand, especially among the Muslim community, is Islamic stocks, which are in accordance with Islamic principles that focus on a healthy economy. This research is focused on predicting Islamic stock prices using the Long Short-Term Memory (LSTM) method and measuring risk with Value at Risk (VaR) using the Cornish-Fisher Expansion (ECF) method. Stock price data from the food sector (PT Indofood), technology sector (Telkom Indonesia), and construction sector (Indocement) for the period 2018-2023 were analyzed. The results show that the ADAM model provides the best performance with the lowest prediction error rates for INTP and TLKM stocks (around 1.22%, 1.98%, and 1.41%). In addition, the SGD model shows limitations in accurate predictions with an error rate above 12%. VaR analysis reveals a slightly higher level of risk in INTP stocks, with a VaR value of around 2.85% at the 95% confidence level. Meanwhile, TLKM stock shows a lower level of risk, with a VaR of around 2.25% at the same confidence level. An in-depth understanding of the risk and growth characteristics of each stock, as well as the selection of the optimization model, are key in making wise investment decisions.

Social Media Analysis and Topic Modeling: Case Study of Stunting in Indonesia Muhaimin, Amri; Fahrudin, Tresna Maulana; Alamiyah, Syifa Syarifah; Arviani, Heidy; Kusuma, Ade; Sari, Allan Ruhui Fatmah; Lisanthoni, Angela
Telematika Vol 20, No 3 (2023): Edisi Oktober 2023
Publisher : Jurusan Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31315/telematika.v20i3.10797

Purpose: Stunting is a problem that currently requires special attention in Indonesia. The stunting rate in 2022 will drop to 21.6%, and for the future, the government has set a target of up to 14% in 2024. Rapid technological developments and freedom of expression on the internet produce review text data that can be analyzed for evaluation. This study analyzes the text data of Twitter users' reviews on stunting. The method used is a text-mining approach and topic modeling based on Latent Dirichlet Allocation.Design/methodology/approach: The methodology used in this study is Latent Dirichlet Allocation. The data was collected from twitter with the keyword 'stunting'. After, the data was cleaned and then modeled using the Latent Dirichlet Allocation.Findings/results: The results show that negative sentiment dominates by 60.6%, positive sentiment by 31.5%, and neutral by 7.9%. In addition, this research shows that 'children', 'decrease', 'number', 'prevention', and 'nutrition' are among the words that often appear on stunting.Originality/value/state of the art: This study uses the keyword stunting and analyzes it. Social media analytics show that the people of Indonesia are primarily aware of stunting. Also, the Latent Dirichlet Analysis can be used to create the model.

Intermittent Data Forecasting using Kernel Support Vector Regression Muhaimin, Amri; Setyowati, Endah; Maulida H, Kartika; Sari, Allan Ruhui Fatma
Nusantara Science and Technology Proceedings 8th International Seminar of Research Month 2023
Publisher : Future Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11594/nstp.2024.4105

Forecasting involves making future estimates. Forecasting methods are commonly employed to predict stock prices, monetary distribution, and weather conditions. To generate accurate forecasts, it is crucial that the data used is consistent, comprehensive, and unchanging. Some data can be readily predicted, while some poses a considerable challenge. An illustration of this is found in discontinuous data, which is notably hard to forecast. Discontinuous data is marked by frequent instances of zero values due to sporadic events. For instance, when tracking the sales of aircraft or other products, sales do not transpire daily, causing recorded data to often register as zero. Various techniques have been explored to handle this kind of data. In this particular study, the chosen method is support vector regression. This method is capable of predicting discontinuous data with a quality level of 1.004, which is lower than traditional approaches like exponential smoothing.

Sentiment Analysis in Social Media: Case Study in Indonesia Muhaimin, Amri; Fahrudin, Tresna Maulana; Alamiyah, Syifa Syarifah; Arviani, Heidy; Kusuma, Ade; Sari, Allan Ruhui Fatmah; Lisanthoni, Angela
Nusantara Science and Technology Proceedings 8th International Seminar of Research Month 2023
Publisher : Future Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11594/nstp.2024.4106

Stunting is a problem that currently requires special attention in Indonesia. The stunting rate in 2022 will drop to 21.6% and for the future, the government has set a target of up to 14% in 2024. There have been many government efforts in implementing programs to reduce stunting rates. However, not everything runs optimally. Rapid technological developments and freedom of expression in the internet world produce review text data that can be analyzed for evaluation. This study aims to analyze the text data of Twitter users' reviews on stunting. The method used is a text-mining approach and topic modeling based on Latent Dirichlet Allocation (LDA). The results show that negative sentiment dominates by 60.6%, positive sentiment by 31.5%, and neutral by 7.9%. In addition, this research shows that 'anak', 'turun', 'angka', 'cegah' and 'gizi' are among the words that often appear on the topic of stunting.

Forecasting The Number of Traffic Accidents in Purbalingga Regency on 2023 Using Time Series Model Trimono; Muhaimin, Amri; Selayanti, Nabilah
Nusantara Science and Technology Proceedings 8th International Seminar of Research Month 2023
Publisher : Future Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11594/nstp.2024.4168

Accident data from Satlantas Purbalingga Regency shows that in 2022 there is an increase in the number of traffic accidents in the Purbalingga Regency. In the future, the impact of accidents is predicted to be bigger so it is necessary to forecasting. Forecasting is one of the most important elements in decision making, because effective or not a decision generally depends on several factors that can not be seen at the time the decision was taken. In this time study the possible time series model is ARMA (2,2), ARMA (2,1), ARMA (1,2), ARMA (1,1), AR (2), AR (1), MA (2), MA (1). However, after testing, the model used is ARMA (1,1). This model is used because it meets all the assumption requirements that are parameter significant, residual independent test, residual normality test, and the smallest Mean Square Error value. According to data forecasting results the highest number of crashes existed in January of 97 accidents and the lowest in December amounted to 93 accidents, So the necessary action from the relevant agencies to cope with the increasing number of traffic accidents in the Purbalingga Regency.

FORECASTING THE OCCUPANCY RATE OF STAR HOTELS IN BALI USING THE XGBOOST AND SVR METHODS Damaliana, Aviolla Terza; Muhaimin, Amri; Prasetya, Dwi Arman
Jurnal Statistika Universitas Muhammadiyah Semarang Vol 12, No 1 (2024): Jurnal Statistika Universitass Muhammadiyah Semarang
Publisher : Department Statistics, Faculty Mathematics and Natural Science, UNIMUS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26714/jsunimus.12.1.2024.24-33

The hotel occupancy rate indicator has become a concern in recent years as it goes hand in hand with the rapid growth of the global tourism industry. A way to maintain or even improve this indicator is to carry out managerial planning using forecasting methods. The forecasting methods used in this research are XGBoost and SVR. The advantage of this modelling is that it achieves high accuracy and processing speed. Meanwhile, the benefit of SVR is that it will produce good prediction because can overcome overfitting. The steps in this research are exploring data, separating training data and testing data, transforming data, modelling data, forecasting data, and evaluating forecasting results using RMSE, MAE, and MAPE. The results show that MAPE value from both methods is smaller than 10%, which means that both methods can predict the occupancy rate of star hotels in Bali very accurately. Apart from that, the SVR method has smaller values for all model evaluation criteria than the XGBoost method, which means that the SVR method is better than XGBoost for predicting the occupancy rate of star hotels in Bali.

Model Selection for Forecasting Rainfall Dataset Muhaimin, Amri; Prabowo, Hendri; Suhartono
IJDASEA (International Journal of Data Science, Engineering, and Analytics) Vol. 1 No. 1 (2021): International Journal of Data Science, Engineering, and Analytics Vol 1, No 1,
Publisher : Universitas Pembangunan Nasional Veteran Jawa Timur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33005/ijdasea.v1i1.2

The objective of this research is to obtain the best method for forecast- ing rainfall in the Wonorejo reservoir in Surabaya. Time series and causal ap- proaches using statistical methods and machine learning will be compared to forecast rainfall. Time series regression (TSR), autoregressive integrated moving average (ARIMA), linear regression (LR), and transfer function (TF) are used as a statistical method. Feedforward neural network (FFNN) and deep feed-for- ward neural network (DFFNN) is used as a machine learning method. Statistical methods are used to capture linear patterns, whereas the machine learning method is used to capture nonlinear patterns. Data about hourly rainfall in the Wonorejo reservoir is used as a case study. The data has a seasonal pattern, i.e. monthly seasonality. Based on the cross-validation and information criteria, the results showed that DFFNN using the time series approach has a more accurate forecast than other methods. In general, machine learning methods have better accuracy than statistical methods. Furthermore, additional information is ob- tained, through this research the parameter that best to make a neural network model is known. Moreover, these results are also not in line with the results of M3 and M4 competition, i.e. more complex methods do not necessarily produce better forecasts than simpler methods.

CLASSIFICATION OF JAVANESE NGLEGENA SCRIPT USING COMPLEXVALUED NEURAL NETWORK Rahmawati, Adinda Aulia; Muhaimin, Amri; Prasetya, Dwi Arman
JIKO (Jurnal Informatika dan Komputer) Vol 7, No 1 (2024)
Publisher : Universitas Khairun

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33387/jiko.v7i1.7808

Javanese script is one of the traditional scripts in Indonesia used by the Javanese people. The Javanese script used in Javanese spelling basically consists of 20 main characters (nglegena), namely from the Ha to Nga script. Javanese script has very high value, the uniqueness of the script is one thing that must be preserved. However, widespread use of Javanese script has declined as technology has developed. In this context, one of the problems that arises is the difficulty in automatically recognizing and classifying the Javanese Nglegena script. Therefore, the use of computational methods to automatically classify the Nglegena Javanese script is very important. This research compares 2 methods for classifying Javanese Nglegena script, namely Complex-Valued Neural Network (CVNN) and Convolutional Neural Network (CNN). This research aims to compare the best accuracy between CVNN and CNN. In this study, the Complex-Valued Neural Network method had a higher average accuracy, namely 96.332% and a loss of 0.1834. Meanwhile, the CNN method has an average accuracy of 93.72% and a loss of 0.4254. Artificial intelligence-based Javanese Nglegena script classification technology can help people to recognize the Javanese Nglegena script, especially in the fields of education and culture.

CLASSIFICATION OF JAVANESE NGLEGENA SCRIPT USING COMPLEXVALUED NEURAL NETWORK Rahmawati, Adinda Aulia; Muhaimin, Amri; Prasetya, Dwi Arman
JIKO (Jurnal Informatika dan Komputer) Vol 7, No 1 (2024)
Publisher : Universitas Khairun

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33387/jiko.v7i1.7808

Javanese script is one of the traditional scripts in Indonesia used by the Javanese people. The Javanese script used in Javanese spelling basically consists of 20 main characters (nglegena), namely from the Ha to Nga script. Javanese script has very high value, the uniqueness of the script is one thing that must be preserved. However, widespread use of Javanese script has declined as technology has developed. In this context, one of the problems that arises is the difficulty in automatically recognizing and classifying the Javanese Nglegena script. Therefore, the use of computational methods to automatically classify the Nglegena Javanese script is very important. This research compares 2 methods for classifying Javanese Nglegena script, namely Complex-Valued Neural Network (CVNN) and Convolutional Neural Network (CNN). This research aims to compare the best accuracy between CVNN and CNN. In this study, the Complex-Valued Neural Network method had a higher average accuracy, namely 96.332% and a loss of 0.1834. Meanwhile, the CNN method has an average accuracy of 93.72% and a loss of 0.4254. Artificial intelligence-based Javanese Nglegena script classification technology can help people to recognize the Javanese Nglegena script, especially in the fields of education and culture.

Title

Found 12 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 12 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 12 Documents
Search