Abstract:The deep forest (DF) model is built using a multilayer ensemble of forest units through decision tree aggregation. DF presents characteristics of an easy-to-understand structure, is suitable for small sample data, and has become an important research direction in the field of deep learning. These attributes are particularly suitable for the modeling of difficult-to-measure parameters in actual industrial process. However, existing methods have mainly focused on the problem of DF classification (DFC) and cannot be directly applied to regression modeling. To overcome these issues, a survey on the DFC algorithm is presented in terms of constructing a small sample data-oriented DF regression (DFR) model for industrial processes. Hence, a survey on the DFC algorithm is presented to construct a small sample size of the data-oriented DF regression (DFR) model for industrial processes. First, principle and properties of DFC are introduced in detail to demonstrate the non-neural network deep learning model. Second, methods of DFC are discussed in terms of feature engineering, representation learning, learner selection, weighting strategy, and hierarchical structure. Furthermore, related studies on decision tree algorithm are reviewed and future investigations on DFR and its relationship with deep learning are discussed and analyzed in detail. Finally, conclusions and the future direction of this study for industrial process modeling are presented. Developing a DFR algorithm with characteristics of dynamic adaptive and interpretation abilities and lightweight structure on the basis of actual industrial domain knowledge will be the focus of our follow-up investigation. Moreover, the existing research results of DFC and deep learning can provide guidance for the future investigations on the DFR model.

A Deep Non-Connectionist Learning Framework for Industrial Data Modeling

An Improved Deep Kernel Partial Least Squares and Its Application to Industrial Data Modeling

DAMPNN: Dynamic Adaptive Message Passing Neural Network for Industrial Soft Sensor

A Deep Supervised Learning Framework Based on Kernel Partial Least Squares for Industrial Soft Sensing.

Deep Residual Principal Component Analysis As Feature Engineering for Industrial Data Analytics

Deep Learning for Industrial KPI Prediction: when Ensemble Learning Meets Semi-Supervised Data

An Efficient Composite Deep Latent Feature Learning Framework With Layerwise Random Mapping for Complicated Industrial Soft Sensor Modeling

Improving Data-Driven Inferential Sensor Modeling by Industrial Knowledge: A Bayesian Perspective

Multivariate Deep Reconstruction Neural Network for Multi-step-ahead Prediction of Industrial Process Quality Variables

A General Knowledge-Guided Framework Based on Deep Probabilistic Network for Enhancing Industrial Process Modeling

Deep Latent Variable Predictive Modeling with Online Bayesian Soft Attention Mechanism

Gated Stacked Target-Related Autoencoder: A Novel Deep Feature Extraction and Layerwise Ensemble Method for Industrial Soft Sensor Application

Self-adaptive Graph Convolution Networks with Application to Industrial Soft Sensor Modeling

An Incremental Deep Convolutional Computation Model for Feature Learning on Industrial Big Data

A Soft Sensor Modeling Framework Embedded with Domain Knowledge Based on Spatio-Temporal Deep LSTM for Process Industry

A Deep Supervised Learning Framework for Data-Driven Soft Sensor Modeling of Industrial Processes

Deep Learning Based Nonlinear Principal Component Analysis for Industrial Process Fault Detection.

Fuzzy Recurrent Stochastic Configuration Networks for Industrial Data Analytics

Anomaly detection using large-scale multimode industrial data: An integration method of nonstationary kernel and autoencoder

Deep Unlstm Network: Features with Memory Information Extracted from Unlabeled Data and Their Application on Industrial Unsupervised Industrial Fault Detection

DF classification algorithm for constructing a small sample size of data-oriented DF regression model