论文信息 - Analysis of WTTE-RNN Variants that Improve Performance

Analysis of WTTE-RNN Variants that Improve Performance

Businesses typically have assets such as machinery, electronics or their customers. These assets share a common trait in that at some stage they will fail or, in the case of customers, they will churn. Knowing when and where to focus limited resources is a key area of concern for businesses. A prediction model called the WTTE-RNN was shown to be effective for predicting the time to event for topics such as machine failure. The purpose of this research is to identify neural network architecture variants of the WTTE-RNN model that have improved performance. The research results on these WTTE-RNN model variant would be useful in the application of the model.

John Burns | Rory Cawley | Rory Cawley | Dr.John Burns | John Burns

[1] Kurt Hornik,et al. Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[2] Benjamin Schrauwen,et al. Training and Analysing Deep Recurrent Neural Networks , 2013, NIPS.

[3] Shih-Chii Liu,et al. Phased LSTM: Accelerating Recurrent Network Training for Long or Event-based Sequences , 2016, NIPS.

[4] Jesper Ljungehed,et al. Predicting Customer Churn Using Recurrent Neural Networks , 2017 .

[5] Roberto Pagliari,et al. Customer Lifetime Value Prediction Using Embeddings , 2017, KDD.

[6] Carlos Guestrin,et al. "Why Should I Trust You?": Explaining the Predictions of Any Classifier , 2016, ArXiv.

[7] Rajib Rana,et al. Gated Recurrent Unit (GRU) for Emotion Classification from Noisy Speech , 2016, ArXiv.

[8] Bin Ran,et al. A hybrid deep learning based traffic flow prediction method and its understanding , 2018 .

[9] Egil Martinsson,et al. WTTE-RNN : Weibull Time To Event Recurrent Neural Network A model for sequential prediction of time-to-event in the case of discrete or continuous censored data, recurrent events or time-varying covariates , 2017 .

[10] Jeff Heaton,et al. Programming Neural Networks with Encog 2 in Java , 2010 .