Interpretable Machine Learning Methods Research Articles

On Leveraging Machine Learning in Sport Science in the Hypothetico-deductive Framework.

Supervised machine learning (ML) offers an exciting suite of algorithms that could benefit research in sport science. In principle, supervised ML approaches were designed for pure prediction, as opposed to explanation, leading to a rise in powerful, but opaque, algorithms. Recently, two subdomains of ML-explainable ML, which allows us to "peek into the black box," and interpretable ML, which encourages using algorithms that are inherently interpretable-have grown in popularity. The increased transparency of these powerful ML algorithms may provide considerable support for the hypothetico-deductive framework, in which hypotheses are generated from prior beliefs and theory, and are assessed against data collected specifically to test that hypothesis. However, this paper shows why ML algorithms are fundamentally different from statistical methods, even when using explainable or interpretable approaches. Translating potential insights from supervised ML algorithms, while in many cases seemingly straightforward, can have unanticipated challenges. While supervised ML cannot be used to replace statistical methods, we propose ways in which the sport sciences community can take advantage of supervised ML in the hypothetico-deductive framework. In this manuscript we argue that supervised machine learning can and should augment our exploratory investigations in sport science, but that leveraging potential insights from supervised ML algorithms should be undertaken with caution. We justify our position through a careful examination of supervised machine learning, and provide a useful analogy to help elucidate our findings. Three case studies are provided to demonstrate how supervised machine learning can be integrated into exploratory analysis. Supervised machine learning should be integrated into the scientific workflow with requisite caution. The approaches described in this paper provide ways to safely leverage the strengths of machine learning-like the flexibility ML algorithms can provide for fitting complex patterns-while avoiding potential pitfalls-at best, like wasted effort and money, and at worst, like misguided clinical recommendations-that may arise when trying to integrate findings from ML algorithms into domain knowledge. KEY POINTS: Some supervised machine learning algorithms and statistical models are used to solve the same problem, y = f(x) + ε, but differ fundamentally in motivation and approach. The hypothetico-deductive framework-in which hypotheses are generated from prior beliefs and theory, and are assessed against data collected specifically to test that hypothesis-is one of the core frameworks comprising the scientific method. In the hypothetico-deductive framework, supervised machine learning can be used in an exploratory capacity. However, it cannot replace the use of statistical methods, even as explainable and interpretable machine learning methods become increasingly popular. Improper use of supervised machine learning in the hypothetico-deductive framework is tantamount to p-value hacking in statistical methods.

Predicting temporomandibular disorders in adults using interpretable machine learning methods: a model development and validation study

IntroductionTemporomandibular disorders (TMD) have a high prevalence and complex etiology. The purpose of this study was to apply a machine learning (ML) approach to identify risk factors for the occurrence of TMD in adults and to develop and validate an interpretable predictive model for the risk of TMD in adults.MethodsA total of 949 adults who underwent oral examinations were enrolled in our study. 5 different ML algorithms were used for model development and comparison, and feature selection was performed by feature importance ranking and feature decreasing methods. Several evaluation indexes, including the area under the receiver-operating-characteristic curve (AUC), were used to compare the predictive performance. The precision-recall curve (PR), calibration curve, and decision curve analysis (DCA) further assessed the accuracy and clinical utility of the model.ResultsThe performance of the random forest (RF) model was the best among the 5 ML models. An interpretable RF model was developed with 7 features (gender, malocclusion, unilateral chewing, chewing hard substances, grinding teeth, clenching teeth, and anxiety). The AUCs of the final model on the training set, internal validation set, and external test set were 0.892, 0.854, and 0.857, respectively. Calibration and DCA curves showed high accuracy and clinical applicability of the model.DiscussionAn efficient and interpretable TMD risk prediction model for adults was successfully developed using the ML method. The model not only has good predictive performance, but also enhances the clinical application value of the model through the SHAP method. This model can provide clinicians with a practical and efficient TMD risk assessment tool that can help them better predict and assess TMD risk in adults, supporting more efficient disease management and targeted medical interventions.

Aerodynamic robustness optimization of aeroengine fan performance based on an interpretable dynamic machine learning method

Aerodynamic robustness optimization of aeroengine fan performance based on an interpretable dynamic machine learning method

Using interpretable machine learning methods to identify the relative importance of lifestyle factors for overweight and obesity in adults: pooled evidence from CHNS and NHANES

BackgroundOverweight and obesity pose a huge burden on individuals and society. While the relationship between lifestyle factors and overweight and obesity is well-established, the relative contribution of specific lifestyle factors remains unclear. To address this gap in the literature, this study utilizes interpretable machine learning methods to identify the relative importance of specific lifestyle factors as predictors of overweight and obesity in adults.MethodsData were obtained from 46,057 adults in the China Health and Nutrition Survey (2004–2011) and the National Health and Nutrition Examination Survey (2007–2014). Basic demographic information, self-reported lifestyle factors, including physical activity, macronutrient intake, tobacco and alcohol consumption, and body weight status were collected. Three machine learning models, namely decision tree, random forest, and gradient-boosting decision tree, were employed to predict body weight status from lifestyle factors. The SHapley Additive exPlanation (SHAP) method was used to interpret the prediction results of the best-performing model by determining the contributions of specific lifestyle factors to the development of overweight and obesity in adults.ResultsThe performance of the gradient-boosting decision tree model outperformed the decision tree and random forest models. Analysis based on the SHAP method indicates that sedentary behavior, alcohol consumption, and protein intake were important lifestyle factors predicting the development of overweight and obesity in adults. The amount of alcohol consumption and time spent sedentary were the strongest predictors of overweight and obesity, respectively. Specifically, sedentary behavior exceeding 28–35 h/week, alcohol consumption of more than 7 cups/week, and protein intake exceeding 80 g/day increased the risk of being predicted as overweight and obese.ConclusionPooled evidence from two nationally representative studies suggests that recognizing demographic differences and emphasizing the relative importance of sedentary behavior, alcohol consumption, and protein intake are beneficial for managing body weight status in adults. The specific risk thresholds for lifestyle factors observed in this study can help inform and guide future research and public health actions.

Interpretable machine learning study of a collector based on combined twisted-tape and wavy-tape inserts

Nowadays, the efficiency of air collectors for solar thermal applications is still low, and many researchers tend to use machine learning to predict and model the performance of thermal systems, but most of the existing machine learning methods are uninterpretable, which poses a challenge for machine learning applications. In this paper, a new collector insert with enhanced heat transfer in the form of a combination of wave and helical twisted bands is firstly designed for performance test experiments using solar air collectors. Then, based on the test data, three mutually interpretable machine learning methods, PDP, ALE, and SHAP, are explored for predictive studies of collector performance. The results show that the average efficiency of the collector with inserted structure increases by 19.71 %, 12.25 %, and 17.53 % at inlet flow rates of 2.1 m/s, 3.3 m/s, and 4.5 m/s, respectively. The highest collector efficiency was achieved with a wave plate length of 360 mm, a helical twist ratio of 5.14, and an inlet flow rate of 4.5 m/s. Understanding how much the input affects the output through the interpretability of SHAP proves the value of interpretable machine learning, which is useful in guiding the modification of the collector structure.

Interpretable Clinical Decision-Making Application for Etiological Diagnosis of Ventricular Tachycardia Based on Machine Learning.

Background: Ventricular tachycardia (VT) can broadly be categorised into ischemic heart disease, non-ischemic structural heart disease, and idiopathic VT. There are few studies related to the application of machine learning for the etiological diagnosis of VT, and the interpretable methods are still in the exploratory stage for clinical decision-making applications. Objectives: The aim is to propose a machine learning model for the etiological diagnosis of VT. Interpretable results based on models are compared with expert knowledge, and interpretable evaluation protocols for clinical decision-making applications are developed. Methods: A total of 1305 VT patient data from 1 January 2013 to 1 September 2023 at the Arrhythmia Centre of Fuwai Hospital were included in the study. Clinical data collected during hospitalisation included demographics, medical history, vital signs, echocardiographic results, and laboratory test outcomes. Results: The XGBoost model demonstrated the best performance in VT etiological diagnosis (precision, recall, and F1 were 88.4%, 88.5%, and 88.4%, respectively). A total of four interpretable machine learning methods applicable to clinical decision-making were evaluated in terms of visualisation, clinical usability, clinical applicability, and efficiency with expert knowledge interpretation. Conclusions: The XGBoost model demonstrated superior performance in the etiological diagnosis of VT, and SHAP and decision tree interpretable methods are more favoured by clinicians for decision-making.

Estimation of compressive strength of concrete with manufactured sand and natural sand using interpretable artificial intelligence

To satisfy the design strength of manufactured sand concrete (MSC) in practical engineering applications, a plethora of geotechnical tests are frequently conducted. An effective approach is imperative to reduce the consumption of labor and resources during these tests. The objective of this paper is to introduce an interpretable machine learning (ML) method to evaluate the compressive strength (CS) of MSC. Firstly, a dataset was established by compiling experimental results from 208 literatures. 3382 data points were selected from the dataset for algorithm training. Recursive Feature Elimination with Cross-Validation (RFECV) was employed to select input parameters. Four algorithms with 12 selected input variables and 1 output variable were constructed to predict CS of MSC using Random Forest (RF), Gradient Boosting Decision Trees (GBDT), eXtreme Gradient Boosting algorithm (XGBoost), and Categorical Boosting (CatBoost). The results show that XGBoost has the highest accuracy and generalization ability (R2=0.934, MAE=3.44, RMSE=5.16, MAPE=0.07). To enhance model transparency, SHapley Additive exPlanations (SHAP) was adopted to explain the underlying predictive mechanisms of ML models. Analyses show that, 1) the cement content, curing time, and water content were the main feature parameters influencing the CS of MSC, 2) the cement content, curing time, and water content have a linear increase, logarithmic increase, and exponential decrease with the CS of MSC, respectively. Partial Dependence Plots (PDP) and Individual Conditional Expectation (ICE) plots were used to further analyze the impacts of these significant influencing factors on the CS of MSC. Additionally, the Local Interpretable Model-Agnostic Explanations (LIME) method was employed to investigate thresholds for various material dosages in MSC containing 5–10 % stone powder. Two typical scenarios were selected for analysis, yielding recommended dosage ranges for concrete of two distinct strengths. Finally, a graphical user interface (GUI) for the CS of MSC has been designed, which might be of great use to material engineers. This provides reference and guidance for concrete engineering practice.

Fast predesign methodology of centrifugal compressor for PEMFCs combining a physics-based loss model and an interpretable machine learning method

For boosted hydrogen fuel cells, the centrifugal compressor greatly affects the system's efficiency, stack power and costs. However, the design of the centrifugal compressor is a lengthy and high-cost process. To tackle this issue, a new method combining a physics-based loss model and a machine learning method was proposed to achieve a fast and more accurate preliminary design step. A high-precision physics-based loss model was developed and validated for data generation. Furthermore, two gradient boosting decision tree (GBDT) models with Bayesian hyperparameter turning were established to construct mapping relationships between geometric parameters and aerodynamic performance. Then, the Sobol sensitivity analysis and Shapley Additive Explanations (SHAP)-based interpretability were used to explore the impact of the geometric parameters on the aerodynamic performances. Results showed that the outlet blade angle, impeller outlet diameter, impeller outlet width, inducer shroud diameter, and inducer blade angle (at the shroud) were the five most sensitive parameters. Additionally, some practical design recommendations were extracted using SHAP-based interpretability. The isentropic efficiencies of two redesign cases with optimal geometric parameters were increased by 2.59% and 1.85% compared with the baseline impellers. The performance prediction was completed within a time of 6 s using the proposed new predesign method. This study represents a significant new attempt to advance the preliminary design methodology of centrifugal compressors.

Projecting Large Fires in the Western US With an Interpretable and Accurate Hybrid Machine Learning Method

AbstractMore frequent and widespread large fires are occurring in the western United States (US), yet reliable methods for predicting these fires, particularly with extended lead times and a high spatial resolution, remain challenging. In this study, we proposed an interpretable and accurate hybrid machine learning (ML) model, that explicitly represented the controls of fuel flammability, fuel availability, and human suppression effects on fires. The model demonstrated notable accuracy with a F1‐score of 0.846 ± 0.012, surpassing process‐driven fire danger indices and four commonly used ML models by up to 40% and 9%, respectively. More importantly, the ML model showed remarkably higher interpretability relative to other ML models. Specifically, by demystifying the “black box” of each ML model using the explainable AI techniques, we identified substantial structural differences across ML fire models, even among those with similar accuracy. The relationships between fires and their drivers, identified by our model, were aligned closer with established fire physical principles. The ML structural discrepancy led to diverse fire predictions and our model predictions exhibited greater consistency with actual fire occurrence. With the highly interpretable and accurate model, we revealed the strong compound effects from multiple climate variables related to evaporative demand, energy release component, temperature, and wind speed, on the dynamics of large fires and megafires in the western US. Our findings highlight the importance of assessing the structural integrity of models in addition to their accuracy. They also underscore the critical need to address the rise in compound climate extremes linked to large wildfires.

Impacts of process parameters on diesel reforming via interpretable machine learning

Diesel reforming is a promising hydrogen production technology used for the clean energy conversion of high-carbon-content fuels. Although the reaction system has been established, predicting the optimal reaction conditions for the system remains challenging. Here, we obtained a set of 675 data points from Aspen Plus simulations and trained regression models to predict the reaction condition ranges that yield the highest hydrogen production in diesel reforming. The ETR model achieved the best predictive performance, with an R2 value of 0.99. Interpretable machine learning methods revealed that temperature is a crucial feature determining the baseline hydrogen yield of the diesel reforming reaction, while the steam-to-carbon ratio is key to enhancing hydrogen yield. Our exploratory study underscores the ability of data-driven ML models to uncover the condition-yield relationship in catalytic diesel reforming for hydrogen production by isolating the effects of individual design parameters, a feat that is difficult to achieve through experimental means.

Deep humoral profiling coupled to interpretable machine learning unveils diagnostic markers and pathophysiology of schistosomiasis.

Schistosomiasis, a highly prevalent parasitic disease, affects more than 200 million people worldwide. Current diagnostics based on parasite egg detection in stool detect infection only at a late stage, and current antibody-based tests cannot distinguish past from current infection. Here, we developed and used a multiplexed antibody profiling platform to obtain a comprehensive repertoire of antihelminth humoral profiles including isotype, subclass, Fc receptor (FcR) binding, and glycosylation profiles of antigen-specific antibodies. Using Essential Regression (ER) and SLIDE, interpretable machine learning methods, we identified latent factors (context-specific groups) that move beyond biomarkers and provide insights into the pathophysiology of different stages of schistosome infection. By comparing profiles of infected and healthy individuals, we identified modules with unique humoral signatures of active disease, including hallmark signatures of parasitic infection such as elevated immunoglobulin G4 (IgG4). However, we also captured previously uncharacterized humoral responses including elevated FcR binding and specific antibody glycoforms in patients with active infection, helping distinguish them from those without active infection but with equivalent antibody titers. This signature was validated in an independent cohort. Our approach also uncovered two distinct endotypes, nonpatent infection and prior infection, in those who were not actively infected. Higher amounts of IgG1 and FcR1/FcR3A binding were also found to be likely protective of the transition from nonpatent to active infection. Overall, we unveiled markers for antibody-based diagnostics and latent factors underlying the pathogenesis of schistosome infection. Our results suggest that selective antigen targeting could be useful in early detection, thus controlling infection severity.

Interpretable machine learning methods to predict the mechanical properties of ABX3 perovskites

This paper proposes the utility of interpretable ensemble learning models for predicting the mechanical properties (bulk, shear and Young moduli) of ABX3 perovskite compounds with the A, B, and X referring to the 3 elements that make the cubic 3-dimensional framework of the perovskite compounds. These models consist of 3 ensemble learning techniques namely CatBoost, Random Forest, and XGBoost. To expand the feature space, robust first-principles density functional theory calculations were used to generate some of the input features, namely elastic constants, density, volume per atom, and ground state energy per atom. The order of the input feature ranking that influences the machine learning (ML) model decisions was then determined. For this, we performed correlation analysis on the multi-dimensional input feature space, suppressed features with high collinearity, and selected features with limited correlation. We trained the three ensemble learning techniques on the desired vectorial input feature representation to predict the mechanical properties. Furthermore, we employed the Shapley Additive Explanations (SHAP) algorithm for analysing the intrinsic decision-making rationality of the ensemble learning models. We measured the performance in the context of the error metrics and coefficient of determination, R2. The results show that XGBoost outperforms other approaches when predicting the shear modulus or Young modulus of the perovskite compounds yielding the least error metrics and the highest R2 value (0.97) in the testing phase. However, both CatBoost and Random Forest outperformed XGBoost when attempting to predict the bulk modulus in the testing phase. The deficiency of the XGBoost in predicting the bulk modulus can be ascribed to an overfitting problem which can occur when the ML model gives accurate predictions for training data but not for test data. Furthermore, the SHAP algorithm provides an insight into the order of feature importance (from highest to lowest). Additionally, we conducted a post-analysis using a holistic ranking to analyse the relative importance of the SHAP feature impact comprehension for the examined ensemble learning techniques. Our findings indicate that the elastic constants are the most important input features influencing the predictive decision of the ensemble learning models.

Assessing the destabilization risk of ecosystems dominated by carbon sequestration based on interpretable machine learning method

Increasing carbon sequestration (CS) in soils and biomass is an important land-based solution in mitigating global warming. Ecosystems provide a wide range of ecosystem services (ESs). The necessity to augment CS may engender alterations in the interrelationships among ESs, thereby heightening the probability of ecosystem destabilization. This study developed a framework that integrates machine learning and interpretable predictions to evaluate the destabilization risk resulting from alterations in ecosystem service relationships dominated by CS. We selected Northeastern China as study area to estimate six ESs and identified areas of destabilization risk among the three services most relevant to CS, including food production (FP), soil retention (SR), and habitat quality (HQ). Subsequently, we compared three machine learning models (random forest, extreme gradient boosting, and support vector machine) and introduced the Shapley additive interpretation (SHAP) method for driving mechanism analysis. The results showed that: (1) CS-FP had 30.28% of its area at destabilization risk and is the most significant ecosystem service pair; (2) Heilongjiang Province was the region with the highest destabilization risk of CS, with CS-FP and CS-SR accounting for 44.76% and 52.89% of all regions, respectively; (3) a non-linear relationship and the presence of threshold features between socio-ecological factors and the prediction of destabilization risk. The study has potential practical value for destabilization risks prevention, while also providing a scientific basis for formulating comprehensive carbon management policies and maintaining ecosystem stability.

Prediction of Post-Treatment Visual Acuity in Age-Related Macular Degeneration Patients With an Interpretable Machine Learning Method.

We evaluated the features predicting visual acuity (VA) after one year in neovascular age-related macular degeneration (nAMD) patients. A total of 527 eyes of 506 patients were included. Machine learning (ML) models were trained to predict VA deterioration beyond a logarithm of the minimum angle of resolution of 1.0 after 1 year based on the sequential addition of multimodal data. BaseM models used clinical data (age, sex, treatment regimen, and VA), SegM models included fluid volumes from optical coherence tomography (OCT) images, and RawM models used probabilities of visual deterioration (hereafter probability) from deep learning classifiers trained on baseline OCT (OCT0) and OCT after three loading doses (OCT3), fluorescein angiography, and indocyanine green angiography. We applied SHapley Additive exPlanations (SHAP) for machine learning model interpretation. The RawM model based on the probability of OCT0 outperformed the SegM model (area under the receiver operating characteristic curve of 0.95 vs. 0.91). Adding probabilities from OCT3, fluorescein angiography, and indocyanine green angiography to RawM showed minimal performance improvement, highlighting the practicality of using raw OCT0 data for predicting visual outcomes. Applied SHapley Additive exPlanations analysis identified VA after 3 months and OCT3 probability values as the most influential features over quantified fluid segments. Integrating multimodal data to create a visual predictive model yielded accurate, interpretable predictions. This approach allowed the identification of crucial factors for predicting VA in patients with nAMD. Interpreting a predictive model for 1-year VA in patients with nAMD from multimodal data allowed us to identify crucial factors for predicting VA.

An interpretable machine learning method for risk stratification of patients with acute coronary syndrome

Risk stratification for major adverse cardiovascular events (MACE) within one year in patients with acute coronary syndrome (ACS) undergoing percutaneous coronary intervention (PCI) remains a challenge. Although several predictive models based on machine learning have emerged, they are difficult to understand. This study aimed to develop a machine learning prediction model that is easy to understand and trustworthy by lay people to assess the risk of MACE in ACS patients undergoing PCI within one year of the procedure. This retrospective cohort study used medical data from 1105 patients to construct a machine-learning model. To ensure thoroughness and multidimensionality of model parsing, Shapley Additive explanations (SHAP) analysis and Local interpretable model-agnostic explanations (LIME) interpretation techniques were used to systematically and deeply interpret the constructed models from a global to a detailed level. The study assessed 12 machine learning methods' prediction models and found that the Random Forest model was the most effective in predicting the risk of MACE in ACS patients after undergoing PCI. The model achieved an AUC value of 0.807 in the validation set, with an accuracy of 0.82, and a stable F1 score of 0.51. SHAP analysis ranked eight key feature variables, such as LVEF, in global importance. The weights of each feature range in the prediction model were revealed using LIME analysis. The machine learning prediction model we developed is capable of accurately predicting the likelihood of patients with ACS experiencing a MACE within one year of surgery.

Exploring the response and prediction of phytoplankton to environmental factors in eutrophic marine areas using interpretable machine learning methods

Coastal marine areas are frequently affected by human activities and face ecological and environmental threats, such as algal blooms and climate change. The community structure of phytoplankton—primary producers in marine ecosystems—is highly sensitive to environmental factors, such as temperature, salinity, and nutrients. However, traditional methods for exploring the relationship between phytoplankton communities and environmental factors in eutrophic marine areas are limited by various factors. Therefore, this study employed interpretable machine learning models, integrating high-dimensional data analysis and complex system modeling, to quantitatively and thoroughly analyze the dynamic relationship between phytoplankton communities and environmental variables in high-frequency samples collected over 53 weeks from eutrophic marine areas. The cell abundance of phytoplankton exhibited a distinct “two-peak pattern” variation. Interpretable machine learning model analysis revealed the dynamic contributions of different environmental factors during changes in the phytoplankton community structure. The results showed that temperature was a key environmental factor that affected phytoplankton growth during peak periods. In addition, the contribution of salinity increased during the second peak in phytoplankton abundance, highlighting its central role in the ecological dynamics of this phase. During green tide outbreaks, particularly in Area 01, the contributions of factors such as temperature and salinity increased, whereas those of phosphates and silicates decreased, indicating that green tide outbreaks substantially altered the nutritional dynamics of the ecosystem. Furthermore, different phytoplankton species, such as Skeletonema costatum, Thalassiosira spp., and Nitzschia spp., exhibit varying responses to environmental factors. Hence, the predictions made using random forest and generalized additive models for phytoplankton cell abundance in two marine areas revealed complex nonlinear relationships between environmental factors, such as temperature, salinity, and phytoplankton abundance.

Identification of key risk factors for venous thromboembolism in urological inpatients based on the Caprini scale and interpretable machine learning methods

PurposeTo identify the key risk factors for venous thromboembolism (VTE) in urological inpatients based on the Caprini scale using an interpretable machine learning method.MethodsVTE risk data of urological inpatients were obtained based on the Caprini scale in the case hospital. Based on the data, the Boruta method was used to further select the key variables from the 37 variables in the Caprini scale. Furthermore, decision rules corresponding to each risk level were generated using the rough set (RS) method. Finally, random forest (RF), support vector machine (SVM), and backpropagation artificial neural network (BPANN) were used to verify the data accuracy and were compared with the RS method.ResultsFollowing the screening, the key risk factors for VTE in urology were “(C1) Age,” “(C2) Minor Surgery planned,” “(C3) Obesity (BMI > 25),” “(C8) Varicose veins,” “(C9) Sepsis (< 1 month),” (C10) “Serious lung disease incl. pneumonia (< 1month) ” (C11) COPD,” “(C16) Other risk,” “(C18) Major surgery (> 45 min),” “(C19) Laparoscopic surgery (> 45 min),” “(C20) Patient confined to bed (> 72 h),” “(C18) Malignancy (present or previous),” “(C23) Central venous access,” “(C31) History of DVT/PE,” “(C32) Other congenital or acquired thrombophilia,” and “(C34) Stroke (< 1 month.” According to the decision rules of different risk levels obtained using the RS method, “(C1) Age,” “(C18) Major surgery (> 45 minutes),” and “(C21) Malignancy (present or previous)” were the main factors influencing mid- and high-risk levels, and some suggestions on VTE prevention were indicated based on these three factors. The average accuracies of the RS, RF, SVM, and BPANN models were 79.5%, 87.9%, 92.6%, and 97.2%, respectively. In addition, BPANN had the highest accuracy, recall, F1-score, and precision.ConclusionsThe RS model achieved poorer accuracy than the other three common machine learning models. However, the RS model provides strong interpretability and allows for the identification of high-risk factors and decision rules influencing high-risk assessments of VTE in urology. This transparency is very important for clinicians in the risk assessment process.

Perception of customer satisfaction and complaints based on BERTopic and interpretable machine learning: evidence from hotels in Xi’an

ABSTRACT This study aims to investigate the topics of customer satisfaction and complaints and explore their role in review sentiment prediction by using deep learning and interpretable machine learning methods. Specifically, this study first uses a BERT-based method for topic clustering and embedding, a factor analysis for topic feature extraction and dimensionality reduction and an oversampling method for the imbalance problem. Further, five types of machine learning methods are utilised to forecast review sentiment. To improve the interpretability of the machine learning model, we employ an interpretable framework to explore the role of topic features in review sentiment prediction. The results indicate that customers’ satisfaction mainly focuses on perfect conference facilities and delicious Chinese cuisine, while customers mainly complain about the quality of service, the attitude of staff and the accessibility of the hotel. In addition, ‘air quality and traffic’ is the most important factor influencing the sentiment in reviews. These findings suggest that hotel managers should focus more on the accommodation experience of customers rather than the value-added experience.

Applying interpretable machine learning in computational biology-pitfalls, recommendations and opportunities for new developments.

Recent advances in machine learning have enabled the development of next-generation predictive models for complex computational biology problems, thereby spurring the use of interpretable machine learning (IML) to unveil biological insights. However, guidelines for using IML in computational biology are generally underdeveloped. We provide an overview of IML methods and evaluation techniques and discuss common pitfalls encountered when applying IML methods to computational biology problems. We also highlight open questions, especially in the era of large language models, and call for collaboration between IML and computational biology researchers.

Interpretable machine learning guided by physical mechanisms reveals drivers of runoff under dynamic land use changes

Human activities continuously impact water balances and cycling in watersheds, making it essential to accurately identify the responses of runoff to dynamic changes in land use types. Although machine learning models demonstrate promise in capturing the intricate interplay between hydrological factors, their “black box” nature makes it challenging to identify the dynamic drivers of runoff. To overcome this challenge, we employed an interpretable machine learning method to inversely deduce the dynamic determinants within hydrological processes. In this study, we analyzed land use changes in the Ningxia section of the middle Yellow River across four periods, laying the foundation for revealing how these changes affect runoff. The sub-watershed attributes and meteorological characteristics generated by the Soil and Water Assessment Tool (SWAT) model were used as input variables of the Extreme Gradient Boosting (XGBoost) model to simulate substantial sub-watershed rainfall runoff in the region. The XGBoost was interpreted using the SHapley Additive exPlanations (SHAP) to identify the dynamic responses of runoff to the land use changes over different periods. The results revealed increasingly frequent interchanges between the land use types in the study area. The XGBoost effectively captured the characteristics of the hydrological processes in the SWAT-derived sub-watersheds. The SHAP analysis results demonstrated that the promoting effect of agricultural land (AGRL) on runoff gradually weakens, while forests (FRST) continuously strengthen their restraining effect on runoff. Relevant land use policies provide empirical support for these findings. Furthermore, the interaction between meteorological variables and land use impacts the runoff generation mechanism and exhibits a threshold effect, with the thresholds for relative humidity (RH), maximum temperature (MaxT), and minimum temperature (MinT) determined to be 0.8, 25 °C, and 15 °C, respectively. This reverse deduction method can reveal hydrological patterns and the mechanisms of interaction between variables, helping to effectively addressing constantly changing human activities and meteorological conditions.