Software Reliability Prediction using Fuzzy Min-Max Algorithm and Recurrent Neural Network Approach

Abstract

Fuzzy Logic (FL) together with Recurrent Neural Network (RNN) is used to predict the software reliability. Fuzzy Min-Max algorithm is used to optimize the number of the kgaussian nodes in the hidden layer and delayed input neurons. The optimized recurrent<br />neural network is used to dynamically reconfigure in real-time as actual software failure. In this work, an enhanced fuzzy min-max algorithm together with recurrent neural network based machine learning technique is explored and a comparative analysis is performed for the modeling of reliability prediction in software systems. The model has been applied on data sets collected across several standard software projects during system testing phase with fault removal. The performance of our proposed approach has been tested using distributed system application failure data set.

Software reliability prediction: A survey

“…Figure 7 shows the classification of these methods. All these methods can use for software reliability prediction 36,39,66,68–71,119,129–139 …”

“…combined fault detection and correction processes and used RNNs to model them and also employed genetic algorithms to optimize the network architecture and to improve fault prediction. Bhuyan and Sethi 69 suggested a feed forward recurrent diffusion network model to forecast reliability using failure data and their results had good correspondence compared to other models. Wang and Zhang 36 have presented an in‐depth RNN based learning model to forecast software reliability.…”

“…In summary, RNNs have been successfully applied to temporal sequence problems. A sample of neural network‐based time series papers from major journals and conferences is reviewed in Table 3 36,39,66,69–71 …”

“…A sample of neural network-based time series papers from major journals and conferences is reviewed in Table 3. 36,39,66,[69][70][71]…”

LSTM Encoder-Decoder Dropout Model in Software Reliability Prediction

“…Wang et al [24] used the Deep Belief Neural Network model to record the semantic characteristics of the program's abstract syntax tree (ASTs) and used this model to predict or detect software faults. Bhuyan et al [25] proposed a detailed feed-forward back propagation network model for predicting reliability using failure data, and the obtained results showed a good fit compared with other models. Jinyong Wang et al [26] suggested a deep learning model based on the recurrent neural network (RNN) Encoder-Decoder for software reliability prediction, the architecture of which uses one / 3 IJRRS/ Vol.…”