Article Per Year (5 Year)
p-Index From 2021 - 2026
1.037
P-Index
This Author published in this journals
All Journal Jurnal Sistem Informasi dan Informatika (SIMIKA) Toba: Journal of Tourism, Hospitality, and Destination Jurnal Kepariwisataan: Destinasi, Hospitalitas dan Perjalanan Jurnal Informatika Buletin Ilmiah Litbang Perdagangan
Erna Nurmawati
Politeknik Statistika STIS
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Environmental Science Languange, Linguistic, Communication & Media Social Sciences

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

 1 
ANALISIS SENTIMEN DAN PEMODELAN TOPIK PADA TWEET TERKAIT DATA BADAN PUSAT STATISTIK Erna Nurmawati; Adielia Amanda
Jurnal Sistem Informasi dan Informatika (Simika) Vol 6 No 2 (2023): Jurnal Sistem Informasi dan Informatika (Simika)
Publisher : Program Studi Sistem Informasi, Universitas Banten Jaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47080/simika.v6i2.2789

Abstract

Annually, the Central Bureau of Statistics, known in Indonesia as BPS or Badan Pusat Statistik, conducts a routine Data Needs Survey (Survei Kebutuhan Data or SKD) to identify data requirements and the level of consumer satisfaction with the quality of data produced by BPS. However, SKD respondents are limited to consumers who have received services from the Integrated Statistics Services (Pelayanan Statistik Terpadu or PST) unit at BPS within a specific year. To gather opinions from the wider public accessing BPS data through channels other than the PST unit, an alternative approach is necessary – particularly through social media, specifically Twitter. This study employs Twitter data to analyze public sentiment regarding BPS data. To understand the distribution of topics discussed within the community about BPS data indicators, topic modeling has been employed. The sentiment analysis process utilizes IndoBERT, an Indonesian language Bidirectional Encoder Representations from Transformers (BERT) model. For topic modeling, the Latent Dirichlet Allocation (LDA) method is utilized. The results of sentiment analysis during the period 2020 - 2022 reveal that tweets related to BPS data generally convey a neutral sentiment. Meanwhile, the topic modeling process generates a range of topics, with variations observed in each year. Throughout 2020 - 2022, the most frequently discussed topics align with the statistical data from the 2020 - 2022 Data Needs Survey's data requirements section, reflecting the diversity of data needs.
Analyzing Instagram Engagement to Forecast Domestic Tourist Trips in Lake Toba and North Sumatra: A Dual Approach with Conventional Statistics and Machine Learning Techniques Erna Nurmawati; Teguh Sugiyarto; Navika Artiari; Adelina Rahmawati
Jurnal Kepariwisataan: Destinasi, Hospitalitas dan Perjalanan Vol. 8 No. 2 (2024)
Publisher : Research and Community Service Center, Politeknik Pariwisata NHI Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34013/jk.v8i2.1619

Abstract

The tourism industry is well known as one booster for economic development. The advance of the tourism industry will lead to the improvement of other economic sectors. Therefore, the Indonesian government is taking steps to ensure the development of its tourism industry by launching 10 super-priority destinations (DSP). Despite numerous efforts and interventions, evidence suggests that the demand for the tourism industry in certain DSPs remains unsatisfied. This also holds true for Lake Toba in North Sumatra. Therefore, it is important to understand how to promote the destination site effectively and increase the number of domestic visitors. This study is aimed at assessing the impact of digital marketing through Instagram to determine the number of domestic tourist trips. The engagement rate (ER) on Instagram posts represents the impact of digital marketing. The result reveals that the topic 'cultural tourism and its activities that develop the economy' has the highest average ER, reaching 692.48. Further analysis reveals that the LSTM model, with independent variables TPK, GTI, and ER on the topic of 'ticket information and vacation packages', is the most effective model for predicting the number of domestic tourist trips to North Sumatra. This analysis emphasizes the crucial role of digital marketing to shape the demand for the tourism industry. The conclusion is based on the significant influence of the Google Trends Index (GTI) and ER on Instagram posts, which serve as a gauge for domestic visitor numbers. The related stakeholders must consider this aspect to sustain its business.
The Impact of Indonesian Textile Imports on Employment: Predictive Analysis with Google Trends and News Sentiment: Politeknik Statistika STIS Dwi Intan Sulistiana; Erna Nurmawati
Buletin Ilmiah Litbang Perdagangan Vol. 19 No. 1 (2025): Buletin Ilmiah Litbang Perdagangan
Publisher : IPB University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29244/bilp.19.1.47-67

Abstract

The textile and textile products (TTP) industry in Indonesia is one of the import-dependent sectors. The increase in imports of the textile industry has the potential to reduce the number of workers. This study aims to identify Harmonized System (HS) codes of TTP import that correlate with the number of workers and to predict imports for those HS codes.  This research employs conventional statistical methods, including Autoregressive Integrated Moving Average (ARIMA), Seasonal ARIMA, ARIMA with Exogenous (ARIMAX), SARIMAX, and Holt-Winters, as well as machine learning methods such as Long Short-Term Memory (LSTM) and Extreme Gradient Boosting (XGBoost), and ARIMA-LSTM hybrid models. The best model is the ARIMAX model, which has the lowest Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). This model utilizes the most influential variables: the rupiah exchange rate, textile production index, percentage of news articles with positive sentiment, and Google Trends Index. This study also reveals that the volume of textile imports, as classified under HS codes 56, 60, and 63, is negatively correlated with the number of workers in the textile sector. Therefore, the government should consider import control policies for this product group. This step needs to be accompanied by an increase in the production capacity and competitiveness of the domestic textile industry. Additionally, the use of Google Trends data and news sentiment can serve as an early warning system to predict import surges more quickly and accurately.
Peran Media Sosial Instagram dalam Memprediksi Jumlah Wisatawan Nusantara (Studi Kasus: Labuan Bajo, Nusa Tenggara Timur) Navika Artiari; Erna Nurmawati; Adelina Rahmawati; Teguh Sugiyarto
TOBA: Journal of Tourism, Hospitality, and Destination Vol. 5 No. 2 (2026): May 2026
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/toba.v5i2.7577

Abstract

The use of big data derived from social media platforms such as Instagram has increasingly been adopted as an alternative data source in tourism analysis, particularly to understand destination promotion patterns and online audience engagement. This study aims to identify the main topics of tourism promotional content on social media and to examine their contribution to predicting the number of domestic tourist trips to the Province of East Nusa Tenggara. Topic modeling was conducted using the Latent Dirichlet Allocation (LDA) method on Instagram captions posted by tourism stakeholders in East Nusa Tenggara. The results of the topic modeling revealed three main topics, namely marine tourism attractiveness, tour package and sailing activity offerings, and trip and boat rental promotions. The analysis of engagement rates (ER) indicates that the average monthly ER of tourism-related accounts in East Nusa Tenggara reached 28.07. Among the identified topics, the marine tourism attractiveness topic recorded the highest average ER, at 42.24, indicating strong audience interest in visual content and narratives related to marine natural attractions. Furthermore, the prediction of domestic tourist trips was carried out by integrating macroeconomic variables, tourism indicators, and social media–based variables. The evaluation results demonstrate that the Long Short-Term Memory (LSTM) model incorporating all explanatory variables—including hotel occupancy rates, inflation, transportation price indices, the COVID-19 pandemic variable, account-level ER, and topic-specific ER—achieved the best performance. This model produced a Mean Absolute Percentage Error (MAPE) of 18.75 percent, a Mean Absolute Error (MAE) of 83,577.40, and a Root Mean Square Error (RMSE) of 111,630.97.
Predicting Stock Price Movements with Technical, Fundamental, and Sentiment Analysis Using the LSTM Model Muhammad Ighfar Saputra; Erna Nurmawati; Rayhan Abyasa
Jurnal Informatika Vol. 12 No. 1 (2025): April
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/informatika.v12i1.12248

Abstract

The challenge of minimizing risk and maximizing profit is what traders in the stock market have been endeavoring to solve for years. Stock prices typically exhibit the characteristic of volatility, influenced by various factors and necessitate a substantial amount of data to identify patterns in price movements. Considering the significant data requirements and the rapid advancement of big data and artificial intelligence, the LSTM (Long-Short Term Memory) model stands as a suitable approach for utilization in Deep Learning. The independent variables employed encompass technical indicator variables, currency exchange rates, interest rates, the Jakarta Composite Index (IHSG), and sentiment data extracted from Twitter tweets. The results indicate that sentiment analysis using the IndoBERT model achieved an accuracy of 0.69, while LSTM analysis produced the model with the smallest error for the fourth (4th) combination of variables, comprising closing price, technical indicators, IHSG, exchange rate, and Twitter sentiment, as well as the twelfth (12th) combination of variables, encompassing closing price, technical indicators, and IHSG. These combinations yielded average RMSE errors of 1.765E-04 and 1.978E-04, respectively. Hyperparameter optimization is done to six hyperparameter, number of unit layer, dropout rate, learning rate, batch size, optimizer, and timestamps. Following hyperparameter optimization, the best-identified model was the fourth (4th) combination of variables, yielding a minimal error of 7.580E-05 and an RMSE of 332.66 in the evaluation of test data. 
Co-Authors Adelina Rahmawati Adelina Rahmawati Adielia Amanda Dwi Intan Sulistiana Muhammad Ighfar Saputra Navika Artiari Navika Artiari Rayhan Abyasa Teguh Sugiyarto Teguh Sugiyarto