DBC-Forest: Deep forest with binning confidence screening

High temporal and spatial resolution PM2.5 dataset acquisition and pollution assessment based on FY-4A TOAR data and deep forest model in China

The long non-coding RNAs (lncRNAs) are subject of intensive recent studies due to its association with various human diseases. It is desirable to build the artificial intelligence-based models for prediction of diseases or tissues based on the lncRNAs data, which will be useful in disease diagnosis and therapy. The accuracy and robustness of existing models based on the machine learning techniques are subject to further improvement. In this study, we propose a deep learning model, called Multi-Label Classifications with Deep Forest, termed MLCDForest, to address multi-label classification on tissue prediction for a given lncRNA, which can be regarded as an implementation of the deep forest model in multi-label classification. The MLCDForest is a sequential multi-label-grained scanning method, which distinguishes from the standard deep forest model. It is proposed to train in sequential of multi-labels with label correlation considered. A systematic comparison using the lncRNA-disease association datasets demonstrates that our method consistently shows superior performance over the state-of-the-art methods in disease prediction. Considering label correlation in the sequential multi-label-grained scanning, our model provides a powerful tool to make multi-label classification and tissue prediction based on given lncRNAs.

Coronavirus disease-19 (COVID-19) has rapidly spread all over the world. It is found that the low sensitivity of reverse transcription-polymerase chain reaction (RT-PCR) examinations during the early stage of COVID-19 disease. Thus, efficient models are desirable for early-stage testing of COVID-19 infected patients. Chest X-ray (CXR) images of COVID-19 infected patients have shown some bilateral changes. In this paper, deep transfer learning and a deep forest-based model are proposed to diagnose COVID-19 infection from CXR images. Initially, features of X-ray images are extracted using the well-known deep transfer learning model (i.e., ResNet101), which does not require tuning many parameters compared to the deep convolutional neural network (CNN). After that, the deep forest model is utilised to predict COVID-19 infected patients. The deep forest is based upon ensemble learning and requires a small number of hyper-parameters. Additionally, the proposed model is trained on a multi-class dataset that contains four different classes as COVID-19 (+), pneumonia, tuberculosis, and healthy patients. The comparisons are drawn among the proposed deep transfer learning and deep forest-based models, the competitive models. The obtained results show that the proposed model effectively diagnoses COVID-19 infection with an accuracy of 99.4%.

With the rapid advancement of DNA sequencing, metagenomics and metatranscriptomics have made great progress, which deepen our understanding on the human microbiome and its impact on human health and diseases. The microbiome, which is characterized by small samples, high dimensions and complicated relationships with hosts, refers to the species, genes and genomes of the microbiota, as well as the products of the microbiota and the host environment. In fact, many machine learning methods have been used to conduct Microbiome-Wide Association Studies which can link the microbiome with the phenotypes, such as the status of human health and diseases. However, existing methods such as Support Vector Machines (SVMs) have some limitations on deep representation learning with deep architectures which can promote the reuse of features and potentially lead to progressively more abstract features at higher layers of representations. Recently, Deep Neural Networks (DNNs), a kind of deep learning models, are widely used for metagenomic data analysis and can perform well on representation learning. But they are considered as a black box and sufferring from criticisms due to theirs lacking of interpretability. Thus, it is interesting to explore other deep learning models for metagenomic data analysis. In this work, we introduce a deep learning model called Deep Forest to study the microbiome associations and we also present an ensemble method for feature selection. Experimental results show that Deep Forest outperforms the traditional machine learning methods. In addition, compared to DNNs, Deep Forest has better interpretability and less hyperparameters.

With the continuous development of digital technology and the continuous improvement of medical information databases, effective mining methods for potentially useful information behind medical data have become one of the research focuses of digital development in the medical field. In this paper, based on the deep forest model, a deep forest classifier framework based on the federated learning algorithm is constructed for the privacy protection of medical data and optimized by gradient boosting decision tree. Then the feature vector of cardiovascular diseases is constructed from engineering features, and feature selection is carried out through the constructed classifier algorithm to realize the prediction and diagnosis of epidemic diseases, followed by the experimental analysis of the method of this paper. The empirical analysis shows that the diagnostic accuracy of this paper’s model in seven common cardiovascular diseases is higher than 80%. Among them, the accuracy rate of heart valve disease is as high as 87%, and the diagnostic accuracy rate of arrhythmia and coronary heart disease is tied for second place with 83%. It shows that the predictive diagnosis model based on deep forest in this paper has good performance, can meet the actual needs of predictive diagnosis of cardiovascular diseases, and provides an effective reference for the development of auxiliary diagnosis in the digital era.

Accurate prediction of wind speed is a prerequisite for the safe and accurate operation of wind power generation, however, WRF models typically do not produce sufficiently accurate wind speed predictions. This study proposed a Seasonal and Temporal Correlation - Deep Forest (STC-DF) model for offshore wind speed prediction. Different from traditional methods, the STC-DF model takes the advantages of the deep forest algorithm to automatically learn complex feature interactions without manual feature engineering. The model is designed to capture the seasonal and temporal characteristics of wind speed variations. To test the effectiveness of the proposed method, we applied the trained STC-DF model to an offshore wind farm in Hainan Province, China. Seven days of data from each season were selected for testing. The results show that the STC-DF model can effectively reduce the error caused by WRF forecast. The error index of the corrected wind speed reduced more than 40%, the accuracy of wind speed forecast increased 15%. And the method passed the multi-model comparison test and robustness experiment. These research results show that the STC-DF model has strong versatility and good correction ability, and is suitable for wind speed forecasting in different regions, which is a feasible method to improve the reliability of offshore wind power generation.

The continuous rise of traffic encryption in IoT devices has led network operators to revisit the way they gain visibility into the behavior of their network and connected assets. Moreover, flow-level analysis is perceived as a more cost-effective approach in network monitoring, particularly at scale, given the high computing cost of deep packet inspection engines. This paper uses time-series signals captured from the flow activity of IoT devices and classifies network traffic with deep learning-based classifiers based on Neural Networks (NN) and Decision Trees (DT). We analyze the efficiency and efficacy of deep learning models using one-dimensional convolutional neural networks (1D-CNN), Long Short Term Memory (LSTM), and Deep Forest (DF). We train our models on the real network traffic of 10 IoT devices collected from our lab during two months. To the best of our knowledge, this study is the first to investigate the performance of DF classifiers on IoT network traffic data and compare them to deep neural network models. We quantify the performance of our models by varying the window size (one minute to three minutes) in a time-series format. We show that the DF models present similar performance to 1D-CNN and LSTM and outperform the (shallow) Random Forest (RF) model but significantly higher inference time. DFs are attractive models since they have a dynamic architecture adjusted during training. Therefore, there is no need to manually search for the model architecture required for deep neural networks.

Machine and deep learning techniques have received increasing attentions in estimating finger forces from high-density surface electromyography (HDsEMG), especially for neural interfacing. However, most machine learning models are normally employed as block-box modules. Additionally, most previous models suffer from performance degradation when dealing with noisy signals. In this work, we propose to employ a forest ensemble model for HDsEMG-force modeling. Our model is explainable and robust against noise. Additionally, we explored the effect of increasing the depth of forest models in EMG-force modeling problems. We evaluated the performance of deep forests with a finger force estimation task. Training and testing data were acquired 3-25 days apart, approximating realistic scenarios. Results showed that deep forests significantly outperformed other models. With artificial signal distortion in 20% channels, deep forests also showed a higher robustness, with the error reduced from that of the baseline by 50% compared with all other models. We provided explanations for the proposed model using the mean decrease impurity (MDI) metric, revealing a strong correspondence between the model and physiology.

In order to further improve the accuracy and efficiency of detecting IoT-Botnet attacks from massive and high-dimensional IoT traffic data with seriously imbalanced class distribution, a botnet traffic detection method based on deep forest is proposed. Firstly, a few dominant features are selected based on the Fisher Score. Then, the multi-grained scanning structure is used to mine more relevant information among low-dimensional features to achieve the purpose of feature enhancement. Finally, the cascade forest performs representation learning and refines the classification layer by layer until the overall performance of the deep forest model converges to the best. The deep forest model is trained and tested on the N_BaIoT dataset, and the results show the deep forest model is not affected by massive and high-dimensional data, its average detection accuracy can reach 99.98% on the imbalanced dataset by using only 6 statistical features.

Syndrome differentiation and treatment is the basic principle of traditional Chinese medicine (TCM) to recognize and treat diseases. Accurate syndrome differentiation can provide a reliable basis for treatment, therefore, establishing a scientific intelligent syndrome differentiation method is of great significance to the modernization of TCM. With the development of biomdical text mining technology, TCM has entered the era of intelligence that based on data, and model training increasingly relies on the large-scale labeled data. However, it is difficult to form a large standard data set in the field of TCM due to the low degree of standardization of TCM data collection and the privacy protection of patients’ medical records. To solve the above problem, a multi-label deep forest model based on an improved multi-label ReliefF feature selection algorithm, ML-PRDF, is proposed to enhance the representativeness of features within the model, express the original information with fewer features, and achieve optimal classification accuracy, while alleviating the problem of high data processing cost of deep forest models and achieving effective TCM discriminative analysis under small samples. The results show that the proposed model finally outperforms other multi-label classification models in terms of multi-label evaluation criteria, and has higher accuracy in the TCM syndrome differentiation problem compared with the traditional multi-label deep forest, and the comparative study shows that the use of PCC-MLRF algorithm for feature selection can better select representative features.

This paper seeks to harness the potential of machine learning techniques to substantially enhance both the precision and efficacy of predicting water hammer phenomena, while simultaneously reducing time, safeguarding pipeline, and conserving resources. A novel mathematical model is proposed by integrating the two-fluid model and the interphase relaxation model with an improved Godunov–Harten–Lax–van Leer numerical method. The study conducts a comprehensive evaluation of the performance of four prominent machine learning algorithms—Back Propagation Neural Network (BP neural network), deep forest, Long Short-Term Memory (LSTM), and Extreme Gradient Boosting (XGBoost)—for the regression analysis of water hammer with column separation. In terms of predictive accuracy, the LSTM model achieves an impressive accuracy of 0.981 94. The BP neural network, deep forest, and XGBoost models yield accuracies of 0.975 01, 0.929 83, and 0.928 65, respectively. Regarding computational efficiency, XGBoost shows a clear advantage, with an overall average execution time of 4.13 s. Deep forest is distinguished by its simplicity in parameter configuration. This research provides valuable insights for selecting the optimal model for regression analysis of water hammer with column separation under a variety of conditions. Furthermore, the paper employs data visualization to directly generate visual trend representations of the overall pressure distribution, thereby eliminating the need for cumbersome numerical comparisons typical of traditional methods.

An unknow fault diagnosis Scheme: A novel random deep forest for fault diagnosis of HVACs

Long noncoding RNAs (lncRNAs) regulate many biological processes by interacting with corresponding RNA-binding proteins. The identification of lncRNA–protein Interactions (LPIs) is significantly important to well characterize the biological functions and mechanisms of lncRNAs. Existing computational methods have been effectively applied to LPI prediction. However, the majority of them were evaluated only on one LPI dataset, thereby resulting in prediction bias. More importantly, part of models did not discover possible LPIs for new lncRNAs (or proteins). In addition, the prediction performance remains limited. To solve with the above problems, in this study, we develop a Deep Forest-based LPI prediction method (LPIDF). First, five LPI datasets are obtained and the corresponding sequence information of lncRNAs and proteins are collected. Second, features of lncRNAs and proteins are constructed based on four-nucleotide composition and BioSeq2vec with encoder-decoder structure, respectively. Finally, a deep forest model with cascade forest structure is developed to find new LPIs. We compare LPIDF with four classical association prediction models based on three fivefold cross validations on lncRNAs, proteins, and LPIs. LPIDF obtains better average AUCs of 0.9012, 0.6937 and 0.9457, and the best average AUPRs of 0.9022, 0.6860, and 0.9382, respectively, for the three CVs, significantly outperforming other methods. The results show that the lncRNA FTX may interact with the protein P35637 and needs further validation.

Deep Forest is a prominent machine learning algorithm known for its high accuracy in forecasting. Compared with deep neural networks, Deep Forest has almost no multiplication operations and has better performance on small datasets. However, due to the deep structure and large forest quantity, it suffers from large amounts of calculation and memory consumption. In this paper, an efficient hardware accelerator is proposed for deep forest models, which is also the first work to implement Deep Forest on FPGA. Firstly, a delicate node computing unit (NCU) is designed to improve inference speed. Secondly, based on NCU, an efficient architecture and an adaptive dataflow are proposed, in order to alleviate the problem of node computing imbalance in the classification process. Moreover, an optimized storage scheme in this design also improves hardware utilization and power efficiency. The proposed design is implemented on an FPGA board, Intel Stratix V, and it is evaluated by two typical datasets, ADULT and Face Mask Detection. The experimental results show that the proposed design can achieve around $40 \times$ speedup compared to that on a 40 cores high performance x86 CPU.

Despite the latest prevailing success of deep neural networks (DNNs), several concerns have been raised against their usage, including the lack of intepretability the gap between DNNs and other well-established machine learning models, and the growingly expensive computational costs. A number of recent works [1],[2],[3] explored the alternative to sequentially stacking decision tree/random forest building blocks in a purely feed-forward way, with no need of back propagation. Since decision trees enjoy inherent reasoning transparency, such deep forest models can also facilitate the understanding of the internal decision making process. This paper further extends the deep forest idea in several important aspects. Firstly, we employ a probabilistic tree whose nodes make probabilistic routing decisions, a.k.a., "soft routing", rather than hard binary decisions. Besides enhancing the flexibility, it also enables non-greedy optimization for each tree. Second, we propose an innovative topology learning strategy: every node in the ree now maintains a new learnable hyperparameter indicating the probability that it will be a leaf node. In that way, the tree will jointly optimize both its parameters and the tree topology during training. Experiments on the MNIST dataset demonstrate that our empowered deep forests can achieve better or comparable performance than [1],[3], with dramatically reduced model complexity. For example, our model with only 1 layer of 15 trees can perform comparably with the model in [3] with 2 layers of 2000 trees each.