Deep Forest Research Articles

The Fault Diagnosis of a Switch Machine Based on Deep Random Forest Fusion

As the key equipment for train operation, the switch machine plays a vital role in the safe and punctual operation of the trains. Nowadays, the fault diagnosis methods of switch machine turnout are mostly based on single-source data. However, it is difficult to fully characterize the fault characteristics using single-source data. In this article, a deep random forest fusion (DRFF) method is proposed to fuse the vibration signals in three directions of the switch machine, which can effectively improve the fault diagnosis accuracy of the switch machine. The fault features are extracted by the wavelet transform method. Subsequently, the features are further optimized by the deep Boltzmann machine. Meanwhile, the DRFF model is formed by using the RFF method to fuse the 3D vibration signals at the feature level. Compared with single-source data and other methods, it is proved that the diagnosis accuracy of the proposed method (98.13%) is far higher than that of other methods, indicating the feasibility of the proposed method, which can greatly improve the fault diagnosis accuracy of the switch machine.

Deep Neural Forest for Out-of-Distribution Detection of Skin Lesion Images.

Deep learning methods have shown outstanding potential in dermatology for skin lesion detection and identification, however, they usually require annotations beforehand and can only classify lesion classes seen in the training set. Moreover, large-scale, open-sourced medical datasets normally have far fewer annotated classes than in real life, further aggravating the problem. This paper proposes a novel method called DNF-OOD, which applies a non-parametric deep forest-based approach to the problem of out-of-distribution (OOD) detection. By leveraging a maximum probabilistic routing strategy and over-confidence penalty term, the proposed method can achieve better performance on the task of detecting OOD skin lesion images, which is challenging due to the large intra-class variability in such images. We evaluate our OOD detection method on images from two large, publicly-available skin lesion datasets, ISIC2019 and DermNet, and compare it against recently-proposed approaches. Results demonstrate the potential of our DNF-OOD framework for detecting OOD skin images.

Multiview Deep Forest for Overall Survival Prediction in Cancer.

Overall survival (OS) in cancer is crucial for cancer treatment. Many machine learning methods have been applied to predict OS, but there are still the challenges of dealing with multiview data and overfitting. To overcome these problems, we propose a multiview deep forest (MVDF) in this paper. MVDF can learn the features of each view and fuse them with integrated learning and multiple kernel learning. Then, a gradient boost forest based on the information bottleneck theory is proposed to reduce redundant information and avoid overfitting. In addition, a pruning strategy for a cascaded forest is used to limit the impact of outlier data. Comprehensive experiments have been carried out on a data set from West China Hospital of Sichuan University and two public data sets. Results have demonstrated that our method outperforms the compared methods in predicting overall survival.

Multiscale Modeling Using GAN and Deep Forest Regression With Application to Dioxin Emission Soft Sensor

Multiscale Modeling Using GAN and Deep Forest Regression With Application to Dioxin Emission Soft Sensor

Predicting Dissolution Kinetics of Tricalcium Silicate Using Deep Learning and Analytical Models

The dissolution kinetics of Portland cement is a critical factor in controlling the hydration reaction and improving the performance of concrete. Tricalcium silicate (C3S), the primary phase in Portland cement, is known to have complex dissolution mechanisms that involve multiple reactions and changes to particle surfaces. As a result, current analytical models are unable to accurately predict the dissolution kinetics of C3S in various solvents when it is undersaturated with respect to the solvent. This paper employs the deep forest (DF) model to predict the dissolution rate of C3S in the undersaturated solvent. The DF model takes into account several variables, including the measurement method (i.e., reactor connected to inductive coupled plasma spectrometer and flow chamber with vertical scanning interferometry), temperature, and physicochemical properties of solvents. Next, the DF model evaluates the influence of each variable on the dissolution rate of C3S, and this information is used to develop a closed-form analytical model that can predict the dissolution rate of C3S. The coefficients and constant of the analytical model are optimized in two scenarios: generic and alkaline solvents. The results show that both the DF and analytical models are able to produce reliable predictions of the dissolution rate of C3S when it is undersaturated and far from equilibrium.

Machine learning-based prediction of infarct size in patients with ST-segment elevation myocardial infarction: A multi-center study

BackgroundCardiac magnetic resonance imaging (CMR) is the gold standard for measuring infarct size (IS). However, this method is expensive and requires a specially trained technologist to administer. We therefore sought to quantify the IS using machine learning (ML) based analysis on clinical features, which is a convenient and cost-effective alternative to CMR. Methods and resultsWe included 315 STEMI patients with CMR examined one week after morbidity in final analysis. After feature selection by XGBoost on fifty-six clinical features, we used five ML algorithms (random forest (RF), light gradient boosting decision machine, deep forest, deep neural network, and stacking) to predict IS with 26 (selected by XGBoost with information gain greater than average level of 56 features) and the top 10 features, during which 5-fold cross-validation were used to train and optimize models. We then evaluated the value of actual and ML-IS for the prediction of adverse remodeling. Our finding indicates that MLs outperform the linear regression in predicting IS. Specifically, the RF with five predictors identified by the exhaustive method performed better than linear regression (LR) with 10 indicators (R2 of RF: 0.8; LR: 0). The finding also shows that both actual and ML-IS were independently associated with adverse remodeling. ML-IS ≥ 21% was associated with a twofold increase in the risk of LV remodeling (P < 0.01) compared with patients with reference IS (1st tertile). ConclusionML-based methods can predict IS with widely available clinical features, which provide a proof-of-concept tool to quantitatively assess acute phase IS.

From multi-omics data to the cancer druggable gene discovery: a novel machine learning-based approach

The development of targeted drugs allows precision medicine in cancer treatment and optimal targeted therapies. Accurate identification of cancer druggable genes helps strengthen the understanding of targeted cancer therapy and promotes precise cancer treatment. However, rare cancer-druggable genes have been found due to the multi-omics data's diversity and complexity. This study proposes deep forest for cancer druggable genes discovery (DF-CAGE), a novel machine learning-based method for cancer-druggable gene discovery. DF-CAGE integrated the somatic mutations, copy number variants, DNA methylation and RNA-Seq data across ˜10 000 TCGA profiles to identify the landscape of the cancer-druggable genes. We found that DF-CAGE discovers the commonalities of currently known cancer-druggable genes from the perspective of multi-omics data and achieved excellent performance on OncoKB, Target and Drugbank data sets. Among the ˜20 000 protein-coding genes, DF-CAGE pinpointed 465 potential cancer-druggable genes. We found that the candidate cancer druggable genes (CDG) are clinically meaningful and divided the CDG into known, reliable and potential gene sets. Finally, we analyzed the omics data's contribution to identifying druggable genes. We found that DF-CAGE reports druggable genes mainly based on the copy number variations (CNVs) data, the gene rearrangements and the mutation rates in the population. These findings may enlighten the future study and development of new drugs.

A Proficient ZESO-DRKFC Model for Smart Grid SCADA Security

Smart grids are complex cyber-physical systems that incorporate smart devices’ communication capabilities into the grid to enable remote management and the control of power systems. However, this integration reveals numerous SCADA system flaws, which could compromise security goals and pose severe cyber threats to the smart grid. In conventional works, various attack detection methodologies are developed to strengthen the security of smart grid SCADA systems. However, they have several issues with complexity, slow training speed, time consumption, and inaccurate prediction outcomes. The purpose of this work is to develop a novel security framework for protecting smart grid SCADA systems against harmful network vulnerabilities or intrusions. Therefore, the proposed work is motivated to develop an intelligent meta-heuristic-based Artificial Intelligence (AI) mechanism for securing IoT-SCADA systems. The proposed framework includes the stages of dataset normalization, Zaire Ebola Search Optimization (ZESO), and Deep Random Kernel Forest Classification (DRKFC). First, the original benchmarking datasets are normalized based on content characterization and category transformation during preprocessing. After that, the ZESO algorithm is deployed to select the most relevant features for increasing the training speed and accuracy of attack detection. Moreover, the DRKFC technique accurately categorizes the normal and attacking data flows based on the optimized feature set. During the evaluation, the performance of the proposed ZESO-DRKFC method is validated and compared in terms of accuracy, detection rate, f1-score, and false acceptance rate. According to the results, it is observed that the ZESO-DRKFC mechanism outperforms other techniques with high accuracy (99%) by precisely spotting intrusions in the smart grid systems.

Multi-label Deep Forest: Towards Automatic Modulation Recognition of Compound Wireless Signals at Low-SNR Environment

Multi-label Deep Forest: Towards Automatic Modulation Recognition of Compound Wireless Signals at Low-SNR Environment

Rapid and Accurate Prediction of Soil Texture Using an Image-Based Deep Learning Autoencoder Convolutional Neural Network Random Forest (DLAC-CNN-RF) Algorithm

Soil determines the degree of water infiltration, crop nutrient absorption, and germination, which in turn affects crop yield and quality. For the efficient planting of agricultural products, the accurate identification of soil texture is necessary. This study proposed a flexible smartphone-based machine vision system using a deep learning autoencoder convolutional neural network random forest (DLAC-CNN-RF) model for soil texture identification. Different image features (color, particle, and texture) were extracted and randomly combined to predict sand, clay, and silt content via RF and DLAC-CNN-RF algorithms. The results show that the proposed DLAC-CNN-RF model has good performance. When the full features were extracted, a very high prediction accuracy for sand (R2 = 0.99), clay (R2 = 0.98), and silt (R2 = 0.98) was realized, which was higher than those frequently obtained by the KNN and VGG16-RF models. The possible mechanism was further discussed. Finally, a graphical user interface was designed and used to accurately predict soil types. This investigation showed that the proposed DLAC-CNN-RF model could be a promising solution to costly and time-consuming laboratory methods.

Fuzzy Deep Forest With Deep Contours Feature for Leaf Cultivar Classification

Deep learning is a compelling technique for feature extraction due to its adaptive capacity of processing and providing deeper image information. However, for the task of leaf cultivar classification, the deep learning-based classifier model is unable to extract contour features of leaf images deeply due to the lack of large specialized datasets and expert knowledge annotations. Also, the scale/size of the current leaf cultivar dataset does not meet the needs of deep neural networks (DNNs). In particular, the high model complexity of DNNs implies that deep-learning-based neural networks seem to must require a large dataset to achieve good performance, but facing the fact that the leaf cultivar dataset often is small, even some classes in this kind of datasets contain less than ten images/examples. To overcome these problems and inspired by the resounding success of fuzzy logic, we propose a novel fuzzy ensemble model for leaf cultivar classification. To extract the contours of leaves, we first propose generative adversarial networks-based methods. Second, to improve the ability of feature representation, we present a data augmentation method to transform our contour features. Third, to get the essential features of leaves, we design a novel generation of the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">fuzzy random forest</i> . Finally, to achieve accurate classification, we design a novel deep learning strategy, namely <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">deep fuzzy representation learning</i> , integrating and cascading a lot of our fuzzy random forests. Experimental results show that our model outperforms other existing state-of-the-arts on three real-world datasets, and performs much better than the original deep forest and DNN-based algorithms particularly.

MRDFF: A deep forest based framework for CT whole heart segmentation.

Automatic whole heart segmentation plays an important role in the treatment and research of cardiovascular diseases. In this paper, we propose an improved Deep Forest framework, named Multi-Resolution Deep Forest Framework (MRDFF), which accomplishes whole heart segmentation in two stages. We extract the heart region by binary classification in the first stage, thus avoiding the class imbalance problem caused by too much background. The results of the first stage are then subdivided in the second stage to obtain accurate cardiac substructures. In addition, we also propose hybrid feature fusion, multi-resolution fusion and multi-scale fusion to further improve the segmentation accuracy. Experiments on the public dataset MM-WHS show that our model can achieve comparable accuracy in about half the training time of neural network models.

Comparison of the efficiency of machine learning algorithms for phishing detection from uniform resource locator

We are using cyberspace for completing our daily life activities because of the growth of Internet. Attackers use some approachs, such as phishing, with the use of false websites to collect personal information of users. Although, software companies launch products to prevent phishing attacks, identifying a webpage as legitimate or phishing, is a very defficult and these products cannot protect from attacks. In this paper, an anti-phishing system has been introduced that can extract feature from website’s URL as instant basis and use four classification algorithms named as K-Nearest neighbor, decision tree, support vector machine, random forest on these features. According to the comparison of the experimental results from these algorithms, random forest algorithm with the selected features gives the highest performance with the 95.67% accuracy rate. Then we have used one deep learning algorithm as enhanced of our experiment named as deep neural decision forests which have given performance with the 92.67% accuracy rate. Then we have created a system which can extract the features from raw URL and pass the features to our deep neural decision forest trained model and can classify the URL as Phishing or legitimate.

Unrest in the Forest and Ethnocide of the Gothi-Koya Tribes in India

The story is of IDP (internally displaced people), particularly of the Gothi-Koya communities, who were compelled to escape from the forests of Chhattisgarh, and are now arbitrarily residing in the deep forests of Telangana State. Their arrival took place specifically during the period of Salva–Judum. Following the debates and demands of human rights activists and intellectuals in connection with the over-violence and illegitimacy of the armed-squad, the Supreme Court of India outlawed it later. The article is based on around a month visit, in the year 2019, to these regions namely Kornapalli, Kristadampadu village of Cherla Block and Pushkunta village of Dammapet Block of Bhadradri Kothagudem District in Telangana. These regions are flanked by dense forests and have been highly sensitive because of their connection with Naxalite movements, and the settlements of the Koyas are scattered, so the researchers had to face various challenges in investigating the issues on a wider scale over there. Researchers made conversation with the displaced people along with the government officials to probe the issue and experienced their plight in their observation.

Improving anti-money laundering in bitcoin using evolving graph convolutions and deep neural decision forest

PurposeMoney laundering is the process of concealing unlawfully obtained funds by presenting them as coming from a legitimate source. Criminals use crypto money laundering to hide the illicit origin of funds using a variety of methods. The most simplified form of bitcoin money laundering leans hard on the fact that transactions made in cryptocurrencies are pseudonymous, but open data gives more power to investigators and enables the crowdsourcing of forensic analysis. With the motive to curb these illegal activities, there exist various rules, policies and technologies collectively known as anti-money laundering (AML) tools. When properly implemented, AML restrictions reduce the negative effects of illegal economic activity while also promoting financial market integrity and stability, but these bear high costs for institutions. The purpose of this work is to motivate the opportunity to reconcile the cause of safety with that of financial inclusion, bearing in mind the limitations of the available data. The authors use the Elliptic dataset; to the best of the authors' knowledge, this is the largest labelled transaction dataset publicly available in any cryptocurrency.Design/methodology/approachAML in bitcoin can be modelled as a node classification task in dynamic networks. In this work, graph convolutional decision forest will be introduced, which combines the potentialities of evolving graph convolutional network and deep neural decision forest (DNDF). This model will be used to classify the unknown transactions in the Elliptic dataset. Additionally, the application of knowledge distillation (KD) over the proposed approach gives finest results compared to all the other experimented techniques.FindingsThe importance of utilising a concatenation between dynamic graph learning and ensemble feature learning is demonstrated in this work. The results show the superiority of the proposed model to classify the illicit transactions in the Elliptic dataset. Experiments also show that the results can be further improved when the system is fine-tuned using a KD framework.Originality/valueExisting works used either ensemble learning or dynamic graph learning to tackle the problem of AML in bitcoin. The proposed model provides a novel view to combine the power of random forest with dynamic graph learning methods. Furthermore, the work also demonstrates the advantage of KD in improving the performance of the whole system.

Igneous rocks lithology identification with deep forest: Case study from eastern sag, Liaohe basin

At present, volcanic reservoir exploration has been carried out widely in the world. In the eastern sag of Liaohe Basin, different volcanic eruption periods lead to complex facies and lithology of igneous rocks, which brings difficulties to logging interpretation. Therefore, a method to accurately identify igneous rock lithology is urgently needed. In this paper, the scheme for identifying igneous rock lithology uses multiple deep forest classifiers. To select a practical training set, 944 conventional logs (AC, DEN, GR, RLLD, and CNL) with lithologic labels were randomly selected from the target area. Compared with the test sets selected from other formations, the accuracy of the scheme reached 99.96%. The scheme was applied to three wells in the same area, and excellent results were achieved when the predicted lithology was compared with the lithology profile and thin section data.

Developing an Improved Survival Prediction Model for Disease Prognosis.

Machine learning has become an important research field in genetics and molecular biology. Survival analysis using machine learning can provide an important computed-aid clinical research scheme for evaluating tumor treatment options. However, the genomic features are high-dimensional, which limits the prediction performance of the survival learning model. Therefore, in this paper, we propose an improved survival prediction model using a deep forest and self-supervised learning. It uses a deep survival forest to perform adaptive learning of high-dimensional genomic data and ensure robustness. In addition, self-supervised learning, as a semi-supervised learning style, is designed to utilize unlabeled samples to improve model performance. Based on four cancer datasets from The Cancer Genome Atlas (TCGA), the experimental results show that our proposed method outperforms four advanced survival analysis methods in terms of the C-index and brier score. The developed prediction model will help doctors rethink patient characteristics' relevance to survival time and personalize treatment decisions.

Application of Improved Deep Learning Method in Intelligent Power System

In view of the inaccurate short-term power load prediction in the power system, where the smart grid cannot effectively coordinate the production, transportation, and distribution of electric energy, the authors propose the application of improved deep learning methods in intelligent power systems. The method uses the convolutional neural network to establish the energy prediction calculation model, uses CNN adaptive data features to mine characteristics, quantifies power uncertainty, uses drop regularization to optimize the deep network structure, uses the deep forest to learn the extracted data features, and builds a prediction model, in order to achieve accurate prediction of power load and solve the problem that the accuracy of existing forecasting methods decreases due to random fluctuations of power. The results showed the following: in the power load forecast results over the weekend, the random forest and the LSTM algorithm forecast results were relatively close and the RMSEs were 17.3 and 17.1, respectively, while the SVM predicted a larger RMSE error of 27.5. The authors’ method predicts the best with 14.8. Conclusion. After verification based on actual load data, in the case of uncertain fluctuations in power load, this method can accurately predict the power load, and the accuracy is higher than that of the more popular methods at present, and it is expected to become an important technical support for solving the core problems of smart grid.

A data-driven fault detection approach for Modular Reconfigurable Flying Array based on the Improved Deep Forest

The Modular Reconfigurable Flying Array, as a kind of special modular rotorcraft UAV, can change its topology configurations fitting for different tasks and varying work scenarios. However, strongly nonlinear characteristics and limited uncertainties of Modular Reconfigurable Flying Array make its sensor faults happen easily and fail to be detected with higher accuracy. In the light of the potential serious losses of its sensors faults, it is valuable and challenging to detect the sensor faults of the aircraft accurately and effectively. Therefore, a data-driven multivariate regression approach based on the Improved Deep Forest is proposed to fulfill sensor faults detection. Firstly, the Deep Forest algorithm is improved by adding the enhanced cascade layer structure and redesigning the inter-layer loss function to pursuit better prediction accuracy. And then, the Improved Deep Forest algorithm is used to establish a multivariate regression model and obtains an estimation of the monitored parameter by learning the historical flight data. Finally, the residual between the actual flight data and the estimated value is calculated to achieve sensor faults detection by comparing with the statistical threshold. What is more, the proposed faults detection approach is evaluated with two real flight datasets collected from the self-made rotorcraft and the experimental results show that the average ACC and AUC are increased by 3% and 2.3% respectively compared with the approach based on the standard Deep Forest.

Small Sample Size and Experience-Independent Hydrate and Pipeline Leakage Identification Technique for Natural Gas Pipelines Based on Deep Forest

Small Sample Size and Experience-Independent Hydrate and Pipeline Leakage Identification Technique for Natural Gas Pipelines Based on Deep Forest