Scholarlink

<p>Accurate prediction of software faults is essential for effective maintenance and improving overall reliability. This study presents a hybrid model that integrates autoregressive integrated moving average (ARIMA) with long short-term memory (LSTM) networks to enhance fault prediction accuracy. The ARIMA part effectively identifies linear patterns and trends in time series data, while the LSTM component captures complex nonlinear relationships and dependencies. Evaluation on three real-world datasets from open-source software projects shows that the hybrid approach outperforms both standalone ARIMA and LSTM models. The advantages of this model include enhanced decision-making capabilities, minimized downtime, and improved user satisfaction. This research provides a significant contribution to the field of software reliability forecasting, offering practitioners a robust tool for ensuring software dependability and enabling proactive strategies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Improving software reliability: a hybrid ARIMA-LSTM approach for fault prediction

  • 期刊论文

  • 发表日期：

    2025年4月1日

查看全文
  • Umashankar Samal,
  • Ajay Kumar

摘要

Accurate prediction of software faults is essential for effective maintenance and improving overall reliability. This study presents a hybrid model that integrates autoregressive integrated moving average (ARIMA) with long short-term memory (LSTM) networks to enhance fault prediction accuracy. The ARIMA part effectively identifies linear patterns and trends in time series data, while the LSTM component captures complex nonlinear relationships and dependencies. Evaluation on three real-world datasets from open-source software projects shows that the hybrid approach outperforms both standalone ARIMA and LSTM models. The advantages of this model include enhanced decision-making capabilities, minimized downtime, and improved user satisfaction. This research provides a significant contribution to the field of software reliability forecasting, offering practitioners a robust tool for ensuring software dependability and enabling proactive strategies.