Article Per Year (5 Year)
p-Index From 2021 - 2026
0.817
P-Index
This Author published in this journals
All Journal bit-Tech International Journal of Data Science, Engineering, and Analytics (IJDASEA)
Mohammad Idhom
UPN "Veteran" Jawa Timur
Computer Science & IT Decision Sciences, Operations Research & Management

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : bit-Tech

 1 
Hybrid Prediction Model Fuzzy Time Series-LSTM on Stock Price Data with Volatility Variation Alfi Hidayatur; Mohammad Idhom; Wahyu Syaifullah
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3014

Abstract

Predicting stock prices in volatile markets remains a major challenge in financial analysis because irregular fluctuations often undermine the reliability of conventional models. Traditional methods such as ARIMA struggle to capture nonlinear dynamics and the complex dependencies that characterize financial time series. To address this gap, this study proposes a hybrid forecasting model that integrates Fuzzy Time Series (FTS) with Long Short-Term Memory (LSTM). The FTS component helps manage uncertainty and simplifies volatility patterns, while the LSTM network captures sequential dependencies across time. Together, these elements provide a more adaptive representation of stock price behavior under different volatility levels. The model was applied to datasets representing both high and low volatility in the Indonesian stock market. Performance was assessed using Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Results show that the hybrid model achieved high accuracy in low-volatility data with an MAE of 284.36 and a MAPE of 0.039%. In high-volatility conditions it also maintained robust performance with an MAE of 885.85 and a MAPE of 0.53%. These outcomes indicate that combining fuzzy logic with deep learning offers a promising approach for stock prediction under volatility variation. The integration not only enhances the reliability of forecasting but also provides a basis for future exploration of risk-aware applications in financial analysis.
A Deep Learning Approach Using Bidirectional-LSTM and Word2Vec for Fake News Classification Fadhilah Nur Hidayat; Wahyu Syaifullah J. S; Mohammad Idhom
bit-Tech Vol. 8 No. 3 (2026): bit-Tech - IN PROGRESS
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i3.3575

Abstract

The rapid growth of online news consumption in Indonesia has intensified the challenge of combating fake news, which undermines public trust and threatens social stability. Conventional approaches, including manual verification, are increasingly inadequate to address the scale and speed of digital information dissemination. This study aims to develop an automatic Indonesian fake news classification system using a deep learning framework that integrates Bidirectional Long Short-Term Memory (Bi-LSTM) with Word2Vec embeddings. Unlike many existing fake news detection models that rely on limited validation settings or focus predominantly on English-language data, this work explicitly addresses the linguistic characteristics and practical constraints of the Indonesian context, thereby strengthening model relevance for real-world deployment. The dataset comprises 6,000 balanced news articles, including 3,000 valid items from Detik.com and 3,000 hoax items from Turnbackhoax.id, collected between January and October 2024. Text preprocessing involved cleaning, stopword removal, tokenization, and padding. A 300-dimensional Word2Vec embedding model was employed, and the classifier was trained using stratified 3-fold cross-validation to ensure robust performance estimation. An ensemble inference strategy was further applied to reduce inter-fold variance and enhance generalization on unseen data, directly addressing a common limitation of prior single-model approaches. Experimental results show that the proposed model achieves an accuracy of 86.43% and an F1-score of 86.28%, alongside a high mean Average Precision of 0.927 during validation. Compared with previously reported deep learning baselines, this framework demonstrates competitive yet more stable performance under realistic evaluation settings, supporting scalable deployment.
Co-Authors Alfi Hidayatur Fadhilah Nur Hidayat Fetty Tri Anggraeny Gideon Setya Budiwitjaksono I Gede Susrama Mas Diyasa Jojok Dwiridotjahjono Munoto Rheza Rizqi Ahmadi Wahyu Syaifullah Wahyu Syaifullah J. S Zainal Abidin Achmad