Deep Forest Research Articles

Blockchain Based Deep Learning for Sustainable Agricultural Supply Chain Management

Food supply chain (FSC) is an important part of the food supply chain. It is essential to establish food supply chains that are open to the public, accountable for their actions, and available in real time. The development of blockchain technology has led to an increase in the amount of data that is passed between customers and businesses, as well as the data that are passed between them. Blockchain technology is a new type of information technology that has the ability to be decentralized, safe, and trusted, which makes it an excellent option for storing sensitive data. The purpose of this study is to evaluate how blockchain with deep learning is used to find the quality evaluation of food supply chain technology. The use of BC has improved the accuracy of food traceability, while the utilization of Deep Random Forest (DRF) has boosted the efficacy of computing and shortened the reaction time. The research compares the quality evaluation system that is based on BC-DRF with some of the most common existing methods in terms of accuracy, reaction time, and sensitivity, when applied to a variety of block sizes.

“Image-Spectral” fusion monitoring of small cotton samples nitrogen content based on improved deep forest

Accurate monitoring of nitrogen nutrition is crucial for improving cotton yield and quality, as well as the ecological environment. The mainstream method for monitoring nutrition is to establish traditional machine learning (ML) models using a single data source. However, this approach has limitations such as limited access to feature information, model-fitting problems, and limited generalization. Deep learning (DL), on the other hand, has shown promise in complex nonlinear modeling tasks due to its flexible structure. However, it has its limitations, such as the fact that agronomic sample collection and testing are usually labor- and material-intensive, resulting in sample sizes that are too small to meet its training conditions. Therefore, there is an urgent need for DL models that can effectively integrate features from multiple data sources and accurately monitor crop nitrogen content, especially in scenarios with small samples. In this study, we conducted indoor pot experiments using the cotton variety Xinluzao 53 and subjected it to six nitrogen treatments. The data sources for our analysis included hyperspectral and digital images of the cotton leaves. To enhance representation learning capabilities, we enriched the multi-class base learners within each layer of the deep forest (DF) model and introduced skip connections. These enhancements improved the quality of inversion for both hyperspectral and digital image datasets. We then developed image-spectral fusion models, which combined the DF structure with stacking ensemble learning. Our focus was on three levels of fusion: feature-level fusion, decision-level fusion, and secondary decision-level fusion. This approach aimed to further enhance the accuracy and stability of nitrogen content inversion. The DF model satisfied the training condition for small samples. Compared to traditional ML algorithms and the original DF algorithm, the improved DF model achieved an increase in validation set R2 of 13.4–28.5% and 10.9–14.9%, respectively. These findings highlight the enhanced accuracy and stability of the improved DF model. Additionally, compared to the optimal inversion model using two single data sources, the “Image-Spectral” three-level fusion models exhibited improvements in validation set R2 of 8.6–9.3%, 10.5–11.2%, and 11.8–12.5% for feature-level, decision-level, and secondary decision-level fusion, respectively. The improved DF and three-level fusion model collectively contributed to the increased accuracy of cotton nitrogen content inversion. Among these models, the secondary decision-level fusion model demonstrated the most marked improvement. This methodology provides valuable insights into monitoring crop phenotypic parameters in situations with limited sample sizes.

Soybean Saponin Content Detection Based on Spectral and Image Information Combination

Soybean saponin is a natural antioxidant and is anti-inflammatory. Hyperspectral analysis technology was applied to detect soybean saponin content rapidly and nondestructively in this paper. Firstly, spectral preprocessing methods were studied, and standard normal variable (SNV) was used to remove noise information. Secondly, a two-step hybrid variable selection approach based on synergy interval partial least squares (SiPLS) and iteratively retains informative variables (IRIV) was proposed to extract characteristic variables. Then, the ensemble learning model was constructed by back propagation neural network (BPNN), deep forest (DF), partial least squares regression (PLSR), and extreme gradient boosting (EXG). Finally, image information was combined into spectral data to improve model accuracy. The prediction coefficient of determination (R2) of the final model reached 0.9216. It can provide rapid, nondestructive, and accurate detection technology of soybean saponin content. A combination of spectral and image information will provide a new idea for application of hyperspectral.

AMPActiPred: A three‐stage framework for predicting antibacterial peptides and activity levels with deep forest

AbstractThe emergence and spread of antibiotic‐resistant bacteria pose a significant public health threat, necessitating the exploration of alternative antibacterial strategies. Antibacterial peptide (ABP) is a kind of antimicrobial peptide (AMP) that has the potential ability to fight against bacteria infection, offering a promising avenue for developing novel therapeutic interventions. This study introduces AMPActiPred, a three‐stage computational framework designed to identify ABPs, characterize their activity against diverse bacterial species, and predict their activity levels. AMPActiPred employed multiple effective peptide descriptors to effectively capture the compositional features and physicochemical properties of peptides. AMPActiPred utilized deep forest architecture, a cascading architecture similar to deep neural networks, capable of effectively processing and exploring original features to enhance predictive performance. In the first stage, AMPActiPred focuses on ABP identification, achieving an Accuracy of 87.6% and an MCC of 0.742 on an elaborate dataset, demonstrating state‐of‐the‐art performance. In the second stage, AMPActiPred achieved an average GMean at 82.8% in identifying ABPs targeting 10 bacterial species, indicating AMPActiPred can achieve balanced predictions regarding the functional activity of ABP across this set of species. In the third stage, AMPActiPred demonstrates robust predictive capabilities for ABP activity levels with an average PCC of 0.722. Furthermore, AMPActiPred exhibits excellent interpretability, elucidating crucial features associated with antibacterial activity. AMPActiPred is the first computational framework capable of predicting targets and activity levels of ABPs. Finally, to facilitate the utilization of AMPActiPred, we have established a user‐friendly web interface deployed at https://awi.cuhk.edu.cn/∼AMPActiPred/.

Enhancing Solar Forecasting Accuracy with Sequential Deep Artificial Neural Network and Hybrid Random Forest and Gradient Boosting Models across Varied Terrains

AbstractEffective solar energy utilization demands improvements in forecasting due to the unpredictable nature of solar irradiance (SI). This study introduces and rigorously tests two innovative forecasting models across different locations: the Sequential Deep Artificial Neural Network (SDANN) and the Deep Hybrid Random Forest Gradient Boosting (RFGB). SDANN, leveraging deep learning, aims to identify complex patterns in weather data, while RFGB, combining Random Forest and Gradient Boosting, proves more effective by offering a superior balance of efficiency and accuracy. The research highlights the SDANN model's deep learning capabilities along with the RFGB model's unique blend and their comparative success over existing models such as eXtreme Gradient Boosting (XGBOOST), Categorical Boosting (CatBOOST), Gated Recurrent Unit (GRU), and a K‐Nearest Neighbors (KNN) and XGBOOST hybrid. With the lowest Mean Squared Error (147.22), Mean Absolute Error (8.77), and a high R2 value (0.80) in a studied region, RFGB stands out. Additionally, detailed ablation studies on meteorological feature impacts on model performance further enhance accuracy and adaptability. By integrating cutting‐edge AI in SI forecasting, this research not only advances the field but also sets the stage for future renewable energy strategies and global policy‐making.

Deep Learning-Based Forest Fire Risk Research on Monitoring and Early Warning Algorithms

With the development of image processing technology and video analysis technology, forest fire monitoring technology based on video recognition is more and more important in the field of forest fire prevention and control. The objects currently applied to forest fire video image monitoring system monitoring are mainly flames and smoke. This paper proposes a forest fire risk monitoring and early warning algorithm, which integrates a deep learning model, infrared monitoring and early warning, and forest fire weather index. The algorithm first obtains the current visible image and infrared image of the same forest area, utilizing a smoke detection model based on deep learning to detect smoke in the visible image, and obtains the confidence level of the occurrence of fire in said visible image. Then, it determines whether the local temperature value of said infrared image exceeds a preset warning value, and obtains a judgment result based on the infrared image. It calculates again a current FWI based on environmental data, and determines a current fire danger level based on the current FWI. Finally, it determines whether or not to carry out a fire warning based on said fire danger level, said confidence level of the occurrence of fire in said visible image, and said judgment result based on the infrared image. The experimental results show that the accuracy of the algorithm in this paper reaches 94.12%, precision is 96.1%, recall is 93.67, and F1-score is 94.87. The algorithm in this paper can improve the accuracy of smoke identification at the early stage of forest fire danger occurrence, especially by excluding the interference caused by clouds, fog, dust, and so on, thus improving the fire danger warning accuracy.

AST-DF: A New Webshell Detection Method Based on Abstract Syntax Tree and Deep Forest

Webshell is a kind of web-language-based website backdoor, which is usually used by attackers to control web servers. Due to its dangerous nature, how to detect Webshell effectively has become a hot research topic in current Web security research. With the rapid development of Webshell evasion technology, the existing Webshell detection methods have the problem of insufficient ability to detect unknown Webshells. In order to solve the above problems and achieve effective Webshell detection, this study proposes a Webshell detection method based on the abstract syntax tree (AST) and deep forest (DF) model called AST-DF. AST-DF first extracts the abstract syntax tree from the PHP code; then, the abstract syntax tree sequence is feature extracted and vectorized using N-gram and TF-IDF. Finally, the vectors are imported into the deep forest model for classification to determine whether the PHP code to be detected is a Webshell or not. The experimental results show that AST-DF achieves remarkable effects in the task of detecting PHP-type Webshells, with a 99.61% accuracy rate, and the values of precision, recall, and F1 score are more than 99%.

DPI_CDF: druggable protein identifier using cascade deep forest

BackgroundDrug targets in living beings perform pivotal roles in the discovery of potential drugs. Conventional wet-lab characterization of drug targets is although accurate but generally expensive, slow, and resource intensive. Therefore, computational methods are highly desirable as an alternative to expedite the large-scale identification of druggable proteins (DPs); however, the existing in silico predictor’s performance is still not satisfactory.MethodsIn this study, we developed a novel deep learning-based model DPI_CDF for predicting DPs based on protein sequence only. DPI_CDF utilizes evolutionary-based (i.e., histograms of oriented gradients for position-specific scoring matrix), physiochemical-based (i.e., component protein sequence representation), and compositional-based (i.e., normalized qualitative characteristic) properties of protein sequence to generate features. Then a hierarchical deep forest model fuses these three encoding schemes to build the proposed model DPI_CDF.ResultsThe empirical outcomes on 10-fold cross-validation demonstrate that the proposed model achieved 99.13 % accuracy and 0.982 of Matthew’s-correlation-coefficient (MCC) on the training dataset. The generalization power of the trained model is further examined on an independent dataset and achieved 95.01% of maximum accuracy and 0.900 MCC. When compared to current state-of-the-art methods, DPI_CDF improves in terms of accuracy by 4.27% and 4.31% on training and testing datasets, respectively. We believe, DPI_CDF will support the research community to identify druggable proteins and escalate the drug discovery process.AvailabilityThe benchmark datasets and source codes are available in GitHub: http://github.com/Muhammad-Arif-NUST/DPI_CDF.

HIERARCHICAL CLUSTERIZATION AND DEEP LEARNING MODEL RANDOM FOREST OF BANKS’ STABILITY UNDER RISK CONDITIONS

Certain theoretical aspects of the stability of Russian banks under risk conditions have been studied. The relevance is due to the fact that in conditions of market uncertainty and risk, approaches to ensure the stability of banks using artificial intelligence are increasingly being used. The goal is to identify patterns between the characteristics of Assets and ROA (Return on Assets), an indicator of return on assets, and obtain a forecast value of Sberbank’s net profit. The result of the study was hierarchical clustering, as well as the generated Deep Learning model Random Forest, which calculated the predicted value of the Sberbank’s net profit. The novelty lies in the fact that the work puts forward and proves the hypothesis that using the Random Forest Deep learning model, a forecast of the net profit of commercial banks can be obtained, which predetermines the stability and dynamics of their development. The conclusions from the study boil down to the fact that a Deep Learning model Random Forest was developed to forecast the amount of net profit, which for Sberbank for 2023 amounted to 38,631 billion rubles, which coincided with its actual value. The area of application of the results obtained is commercial banks.

Forest Cows Secrets: Cracking the Code With Movement Sensors

Have you ever wondered how we can watch animals in the wild without actually being near them? In Norway, cows roam freely in the deep forest during summer. While the cows enjoy the freedom, it can be tricky to keep them safe from carnivores like wolves and bears, as no shepherds or dogs protect the herds. Keeping an eye on the cows is important! Farmers and researchers use GPS to track animals, just as we do for phones or cars. However, GPS does not tell us much about what the animals are doing. That is where movement sensors come in. These sensors store information about the tiniest body movements and reveal what the animal is doing at any time. Is the cow’s head up or down? Is it walking or running? Based on the data, we could distinguish 20 different behaviors! Now we can spy on cows, see what they are up to in the forest, and help farmers better care for them.

Fracture identification of carbonate reservoirs by deep forest model: An example from the D oilfield in Zagros Basin

Identifying fractures along a well trajectory is of immense significance in determining the subsurface fracture network distribution. Typically, conventional logs exhibit responses in fracture zones, and almost all wells have such logs. However, detecting fractures through logging responses can be challenging since the log response intensity is weak and complex. To address this problem, we propose a deep learning model for fracture identification using deep forest, which is based on a cascade structure comprising multi-layer random forests. Deep forest can extract complex nonlinear features of fractures in conventional logs through ensemble learning and deep learning. The proposed approach is tested using a dataset from the Oligocene to Miocene tight carbonate reservoirs in D oilfield, Zagros Basin, Middle East, and eight logs are selected to construct the fracture identification model based on sensitivity analysis of logging curves against fractures. The log package includes the gamma-ray, caliper, density, compensated neutron, acoustic transit time, and shallow, deep, and flushed zone resistivity logs. Experiments have shown that the deep forest obtains high recall and accuracy (>92%). In a blind well test, results from the deep forest learning model have a good correlation with fracture observation from cores. Compared to the random forest method, a widely used ensemble learning method, the proposed deep forest model improves accuracy by approximately 4.6%.

An Extended Feature Representation Technique for Predicting Sequenced-based Host-pathogen Protein-protein Interaction

Background: The use of machine learning models in sequence-based Protein-Protein Interaction prediction typically requires the conversion of amino acid sequences into feature vectors. From the literature, two approaches have been used to achieve this transformation. These are referred to as the Independent Protein Feature (IPF) and Merged Protein Feature (MPF) extraction methods. As observed, studies have predominantly adopted the IPF approach, while others preferred the MPF method, in which host and pathogen sequences are concatenated before feature encoding. Objective: This presents the challenge of determining which approach should be adopted for improved HPPPI prediction. Therefore, this work introduces the Extended Protein Feature (EPF) method. Methods: The proposed method combines the predictive capabilities of IPF and MPF, extracting essential features, handling multicollinearity, and removing features with zero importance. EPF, IPF, and MPF were tested using bacteria, parasite, virus, and plant HPPPI datasets and were deployed to machine learning models, including Random Forest (RF), Support Vector Machine (SVM), Multilayer Perceptron (MLP), Naïve Bayes (NB), Logistic Regression (LR), and Deep Forest (DF). Results: The results indicated that MPF exhibited the lowest performance overall, whereas IPF performed better with decision tree-based models, such as RF and DF. In contrast, EPF demonstrated improved performance with SVM, LR, NB, and MLP and also yielded competitive results with DF and RF. Conclusion: In conclusion, the EPF approach developed in this study exhibits substantial improvements in four out of the six models evaluated. This suggests that EPF offers competitiveness with IPF and is particularly well-suited for traditional machine learning models.

Gaming [the] Systems: Critical Analysis through Tabletop Roleplaying Games in Humanities Classrooms

In its very designation, “role play” contains a productive tension: the expectations bound up in roles and the liberating potentiality of play. It comes as no surprise, then, that role play has found expression in the worlds of business, therapy, education, and even the military. Table-top games exist as our culture’s most familiar form of role play but among its least theorized. In this paper, I examine the rich potential of table-top role-playing games in humanities classrooms as a means to fuel cultural critique, critical analysis, class discussion, and self-reflection. My paper lays out a theoretical structure for analyzing games as multi-dimensional texts and a practical guide for bringing them into the classroom. In particular, I look at two free-to-access and permissively concise game systems which can easily fit into an open week of coursework. The games, “Lasers and Feelings” and “The Deep Forest,” depart from the cultural expectations of “Dungeons & Dragons” to rethink, adapt, and challenge our understanding of issues ranging from the formulation of the rational, Western subject to the politics of post-colonialism. I supplement conventional process-oriented pedagogy with Ian Bogost’s “Persuasive Games” and C. Thai Nguyen’s “Games: Agency as Art,” to establish a robust groundwork for students to analyze and critique how game design decisions make arguments about the world. Furthermore, in playing the games in the classroom and debriefing afterwards, students identify their internal systems that reflect and inform their own assumptions about the world.

Multi-view uncertainty deep forest: An innovative deep forest equipped with uncertainty estimation for drug-induced liver injury prediction

Drug-induced liver injury (DILI) refers to the harmful effects that certain drugs can have on the liver, potentially leading to severe health issues and even life-threatening conditions. Early detection of potential DILI events is crucial in drug development and toxicity assessment to ensure patient safety and the successful market entry of pharmaceuticals. We introduced a novel model, the Multi-view Uncertainty Deep Forest (MVU-DF), for predicting DILI. This model expands the Deep Forest architecture to multiple views and uses two-stage feature construction: intra-view feature interaction for key information within a view, and inter-view feature interaction for sharing information across views. Additionally, we incorporated uncertainty estimation to evaluate the confidence of predictions, using opinion as an alternative to probability. The opinion measures class support and incorporates uncertainty, enabling more precise differentiation of hard and easy samples in each MVU-DF iteration, enhancing the cascading scheme. The model's high interpretability enhances prediction performance and reliability. Our experiments demonstrate the potential of this research in improving the accuracy and decision-making transparency of hepatotoxicity predictions in drug development.

Screening androgen receptor agonists of fish species using machine learning and molecular model in NORMAN water-relevant list

Androgen receptor (AR) agonists have strong endocrine disrupting effects in fish. Most studies mainly investigate AR binding capacity using human AR in vitro. However, there is still few methods to rapidly predict AR agonists in aquatic organisms. This study aimed to screen AR agonists of fish species using machine learning and molecular models in water-relevant list from NORMAN, a network of reference laboratories for monitoring contaminants of emerging concern in the environment. In this study, machine learning approaches (e.g., Deep Forest (DF)), Random Forests and artificial neural networks) were applied to predict AR agonists. Zebrafish, fathead minnow, mosquitofish, medaka fish and grass carp are all important aquatic model organisms widely used to evaluate the toxicity of new pollutants, and the molecular models of ARs from these five fish species were constructed to further screen AR agonists using AlphaFold2. The DF method showed the best performances with 0.99 accuracy, 0.97 sensitivity and 1 precision. The Asn705, Gln711, Arg752, and Thr877 residues in human AR and the corresponding sites in ARs from the five fish species were responsible for agonist binding. Overall, 245 substances were predicted as suspect AR agonists in the five fish species, including, certain glucocorticoids, cholesterol metabolites, and cardiovascular drugs in the NORMAN list. Using machine learning and molecular modeling hybrid methods rapidly and accurately screened AR agonists in fish species, and helping evaluate their ecological risk in fish populations.

An effective deep learning scheme for android malware detection leveraging performance metrics and computational resources

With the rise in the use of Android smartphones, there has been a proportional surge in the proliferation of malicious applications (apps). As mobile phone users are at a heightened risk of data theft, detecting malware on Android devices has emerged as a pressing concern within the realm of cybersecurity. Conventional techniques, such as signature-based routines, are no longer sufficient to safeguard users from the continually evolving sophistication and swift behavioral modifications of novel varieties of Android malware. Hence, there has been a significant drive in recent times towards leveraging machine learning (ML) models and methodologies to identify and generalize malicious behavioral patterns of mobile apps for detecting malware. This paper proposes Deep learning (DL) based on new and highly reliable classifier, deep neural decision forest (DNDF) for detecting Android malware. Two datasets were used: Drebin and 2014 for comparison with previous studies, and TUANDROMD collected in 2021 for detecting the latest threats with advanced obfuscation and morphing techniques. We have also calculated the time-consuming and computational resources taken by our classifier. After conducting a thorough performance evaluation, our proposed approach attained impressive results on two datasets. The empirical findings reveal that the proposed DBN and DNDF models demonstrated exceptional performance, achieving an accuracy of 99%, a sensitivity of 1, and an AUC value of 0.98%. The metrics we obtained are comparable to those of state-of-the-art ML-based Android malware detection techniques and several commercial antivirus engines.

High-spatial-resolution surface soil moisture retrieval using the Deep Forest model in the cloud environment over the Tibetan Plateau

ABSTRACT As a key climate variable, soil moisture plays a crucial role in drought detection, flood warning, and crop yield prediction. In recent years, the demand for high-spatial-resolution soil moisture has increased, particularly in environmental management. In this study, Copernicus Sentinel-1 synthetic aperture radar data, Sentinel-2 multi-spectral data, and other auxiliary data (land cover types, soil texture, etc.) were used to retrieve surface soil moisture (10 m) in the cloud environment (Google Earth Engine + Google Colab + Google Drive) over the Tibetan Plateau, and an entirely data-driven machine learning-based model called Deep Forest was adopted. We discussed the application of the Deep Forest model and compared it with other machine learning models. Overall, on the basis of 10-fold cross-validation, the modified Deep Forest model performed the best, with estimate accuracy of 0.834 and 0.038 m3·m−3 in terms of coefficient of determination ( R 2 ) and unbiased Root Mean Square Error (ubRMSE), respectively. It also demonstrated the best performance in site-based validation ( R 2 of 0.606 and ubRMSE of 0.092 m3·m−3). In addition, the framework for the data acquisition, data preprocessing, model training, and soil moisture mapping in this study was completed in the cloud environment, which facilitated the entire retrieval process. This work provides new ideas beyond the retrieval model for other related studies.

Location-Aware Deep Interaction Forest for Web Service QoS Prediction

With the rapid development of the web service market, the number of web services shows explosive growth. QoS is an important factor in the recommendation scene; how to accurately recommend a high-quality service for users among the massive number of web services has become a tough problem. Previous methods usually acquired feature interaction information by network structures like DNN to improve the QoS prediction accuracy, but this generates unnecessary computations. Aiming at addressing the above problem, inspired by the multigrained scanning mechanism in a deep forest, we propose a location-aware deep interaction forest approach for web service QoS prediction (LDIF). This approach offers the following innovations: The model fuses the location similarity of users and services as a latent feature representation of them. In addition, we designed a scanning interaction structure (SIS), which obtains multiple local feature combinations from the interaction between user and service features, uses interactive computing to extract feature interaction information, and concatenates the feature interaction information with original features, which aims to enhance the dimension of the features. Equipped with these, we compose a layer-by-layer cascade by using SIS to fuse low- and high-order feature interaction information, and the early-stop mechanism controls the cascade depth to avoid unnecessary computation. The experiments demonstrate that our model outperforms eight other state-of-the-art methods on MAE and RMSE common metrics on real public datasets.

TSCF: An Improved Deep Forest Model for Time Series Classification

The deep forest presents a novel approach that yields competitive performance when compared to deep neural networks. Nevertheless, there are limited studies on the application of deep forest to time series classification (TSC) tasks, and the direct use of deep forest cannot effectively capture the relevant characteristics of time series. For that, this paper proposes time series cascade forest (TSCF), a model specifically designed for TSC tasks. TSCF relies on four base classifiers, i.e., random forest, completely random forest, random shapelet forest, and diverse representation canonical interval forest, allowing for feature learning on the original data from three granularities: point, subsequence, and summary statistics calculated based on intervals. The major contribution of this work, is to define an ensemble and deep classifier that significantly outperforms the individual classifiers and the original deep forest. Experimental results show that TSCF outperforms other forest-based algorithms for solving TSC problems.

সুন্দরবনের জল ও জঙ্গলে বাংলাদেশের মুক্তিযুদ্ধ: নব্য-নির্ধারণবাদী বিশ্লেষণ

In the Liberation War of Bangladesh, its geographical features were influential in many ways on the activities of the freedom fighters and their opposing forces. This is even more true in the case of the liberation war of the Sundarbans in water and jungle. In June 1971, Under Lt. Ziauddin's leadership, freedom fighter bases were established inside the Sundarbans. The group is organised in the deep forest, accepting and in some cases overcoming the various obstacles of nature. Using the natural obstacles to their advantage, they again engaged the Pakistani troops and their allied forces in the immediate area. Using their positional advantage, these freedom fighters played an effective role in sending refugees and trainees across the border. On the other hand, the freedom fighters in Sundarbans were used as a transit for many of the freedom fighters who came from July with training and weapons to the interior of the country, especially in the Khulna-Barisal region. This article analyses such activities of the Sundarbans freedom fighters in a neo-deterministic trend.