Interpretive Modeling Research Articles

Capturing key interdependences among supply chain disruptions and mitigation strategies to enhance firm performance

PurposeThis paper aims to capture the complex interdependences between supply chain disruptions (SCDs), SC risk mitigation strategies and firm performance in the context of disruptive events to enhance resilience for medium-sized and large firms coping with complex supply chain networks. The roles of digitalization, insurance and government support have also been addressed as potential strategies to counteract the impacts of disruptions on supply chains.Design/methodology/approachThis study is based on an empirical investigation in an FMCG company – using a hybrid causal mapping technique based on the frameworks of interpretive structural modeling (ISM) and Bayesian networks (BN) – of 11 levels of relationships between SCDs (in supply, production, logistics, demand and finance), SC risk mitigation strategies (flexibility, efficiency, agility and responsiveness), insurance, government support, information and knowledge sharing, digitalization and finally the key firm performance measures (continuity, quality and financial performance).FindingsThe results of the empirical investigation reveal and describe: (1) the nature and probabilistic quantification of the lower-level relationships among the four SCDs, among the mitigation strategies and the three firm performance measures; (2) the nature and probabilistic quantification of the higher-level relationships among the impacts of SCDs, SC risk mitigation strategies and firm performance and (3) how to model and quantify the complex interdependences in single firms and their supply chains.Originality/valueOur results can support managers in developing more effective decision-making models to assess and manage unfavorable events and cascade effects among different functions and processes in the context of risks and disruptions.

Using interpretative structure model and analytical network process for optimum site selection of airport locations in Delta Egypt

Abstract Airports are among the constructions that must meet international standards and specifications established by the International Civil Aviation Organization (ICAO). Some of these parameters and criteria include topography, environmental, and operating circumstances. Navigation operations are also affected by terrain and human barriers, while noise, infrastructure, and weather factors affect the environment around airports. This article investigates using the interpretative structure model (ISM) and analytical network process (ANP) together as a new technique to select and determine the optimum site selection for constructing new international airports in the Nile Delta, Egypt. The criteria used (16 international criteria) are selected depending on the standards of the ICAO, field surveys, and previous studies. Nile Delta, Egypt, is chosen as a case study because it suffers from the lack of international airports and has a population of approximately 23 million. Therefore, new international airports must be established to serve this region, boost tourism, improve transportation, and stimulate commercial traffic. The results using the suggested new technique are compared with the traditional used methods for site selection, such as fuzzy-analytic hierarchy process. Landsat 8 images are used in this research. A quality test using the area under the curve and the receiver operating characteristic curve was applied to evaluate the new technique for site selection of international airports, depending on calculating the highest suitability index for each proposed site. From the quality tests, it is deduced that the suggested method (ISM–ANP) for airport site selection is more accurate than any other traditional method. ArcGIS 10.5 software is used to draw the final digital maps containing the proposed resulted sites. As a result, three new locations for the construction of international airports were found and selected throughout the research region (Delta Nile, Egypt) based on the used mathematical models. Therefore, the proposed novel method for determining the locations of international airports is thought to be effective and feasible, and it can be used to determine the locations of any development projects in general, particularly in developing countries, which benefits the decision-makers in making the right decisions.

An interpretable model for landslide susceptibility assessment based on Optuna hyperparameter optimization and Random Forest

This study proposed an interpretable model that combines Random Forest (RF), Optuna hyperparameter optimization, and SHapley Additive exPlanations (SHAP) to achieve optimal landslide susceptibility evaluation and provide explanations in the northwest region of Yunnan Province in China. First, an inventory of 4447 landslides and 23 related factors was considered for the landslide susceptibility assessment. Subsequently, a hyperparameter-optimized RF model was developed using the Optuna framework and the training dataset to generate landslide susceptibility maps. The performance of the models were evaluated using accuracy (ACC), precision (PPV), recall (TPR), F1-score (F1), and the Area Under the Curve (AUC) based on the Receiver Operating Characteristic. Furthermore, the interpretability of the model was enhanced through the implementation of SHAP. The proposed model demonstrated outstanding performance on the test set, achieving an ACC of 0.7792, PPV of 0.7448, TPR of 0.8769, F1 of 0.8055, and an AUC of 0.8387. The interpretability analysis revealed that elevation, population density, distance from roads, and normalized difference vegetation index were the primary factors influencing landslide occurrences in the study area. This study provides a comprehensive framework for evaluating landslide susceptibility in specific regions and offers invaluable insights for the prevention and management of landslide disasters.

Unlocking the adaptive advantage: correlation and machine learning classification to identify optimal online adaptive stereotactic partial breast candidates.

Online adaptive radiotherapy (OART) is a promising technique for delivering stereotactic accelerated partial breast irradiation (APBI), as lumpectomy cavities vary in location and size between simulation and treatment. However, OART is resource-intensive, increasing planning and treatment times and decreasing machine throughput compared to the standard of care (SOC). Thus, it is pertinent to identify high-yield OART candidates to best allocate resources. 
Approach. Reference plans (plans based on simulation anatomy), SOC plans (reference plans recalculated onto daily anatomy), and daily adaptive plans were analyzed for 31 sequential APBI targets, resulting in the analysis of 333 treatment plans. Spearman correlations between 22 reference plan metrics and 10 adaptive benefits, defined as the difference between mean SOC and delivered metrics, were analyzed to select a univariate predictor of OART benefit. A multivariate logistic regression model was then trained to stratify high- and low-benefit candidates.
Main Results. Adaptively delivered plans showed dosimetric benefit as compared to SOC plans for most plan metrics, although the degree of adaptive benefit varied per patient. The univariate model showed high likelihood for dosimetric adaptive benefit when the reference plan ipsilateral breast V15Gy exceeds 23.5%. Recursive feature elimination identified 5 metrics that predict high-dosimetric-benefit adaptive patients. Using leave-one-out cross validation, the univariate and multivariate models classified targets with 74.2% and 83.9% accuracy, resulting in improvement in per-fraction adaptive benefit between targets identified as high- and low-yield for 7/10 and 8/10 plan metrics, respectively. 
Significance. This retrospective, exploratory study demonstrated that dosimetric benefit can be predicted using only ipsilateral breast V15 Gy on the reference treatment plan, allowing for a simple, interpretable model. Using multivariate logistic regression for adaptive benefit prediction led to increased accuracy at the cost of a more complicated model. This work presents a methodology for clinics wishing to triage OART resource allocation.&#xD.

An interpretable model of pre-mRNA splicing for animal and plant genes.

Pre-mRNA splicing is a fundamental step in gene expression, conserved across eukaryotes, in which the spliceosome recognizes motifs at the 3' and 5' splice sites (SSs), excises introns, and ligates exons. SS recognition and pairing is often influenced by protein splicing factors (SFs) that bind to splicing regulatory elements (SREs). Here, we describe SMsplice, a fully interpretable model of pre-mRNA splicing that combines models of core SS motifs, SREs, and exonic and intronic length preferences. We learn models that predict SS locations with 83 to 86% accuracy in fish, insects, and plants and about 70% in mammals. Learned SRE motifs include both known SF binding motifs and unfamiliar motifs, and both motif classes are supported by genetic analyses. Our comparisons across species highlight similarities between non-mammals, increased reliance on intronic SREs in plant splicing, and a greater reliance on SREs in mammalian splicing.

Prediction of anticancer drug sensitivity using an interpretable model guided by deep learning

BackgroundThe prediction of drug sensitivity plays a crucial role in improving the therapeutic effect of drugs. However, testing the effectiveness of drugs is challenging due to the complex mechanism of drug reactions and the lack of interpretability in most machine learning and deep learning methods. Therefore, it is imperative to establish an interpretable model that receives various cell line and drug feature data to learn drug response mechanisms and achieve stable predictions between available datasets.ResultsThis study proposes a new and interpretable deep learning model, DrugGene, which integrates gene expression, gene mutation, gene copy number variation of cancer cells, and chemical characteristics of anticancer drugs to predict their sensitivity. This model comprises two different branches of neural networks, where the first involves a hierarchical structure of biological subsystems that uses the biological processes of human cells to form a visual neural network (VNN) and an interpretable deep neural network for human cancer cells. DrugGene receives genotype input from the cell line and detects changes in the subsystem states. We also employ a traditional artificial neural network (ANN) to capture the chemical structural features of drugs. DrugGene generates final drug response predictions by combining VNN and ANN and integrating their outputs into a fully connected layer. The experimental results using drug sensitivity data extracted from the Cancer Drug Sensitivity Genome Database and the Cancer Treatment Response Portal v2 reveal that the proposed model is better than existing prediction methods. Therefore, our model achieves higher accuracy, learns the reaction mechanisms between anticancer drugs and cell lines from various features, and interprets the model’s predicted results.ConclusionsOur method utilizes biological pathways to construct neural networks, which can use genotypes to monitor changes in the state of network subsystems, thereby interpreting the prediction results in the model and achieving satisfactory prediction accuracy. This will help explore new directions in cancer treatment. More available code resources can be downloaded for free from GitHub (https://github.com/pangweixiong/DrugGene).

Decoupling the multi-drivers of urban extreme heat environment in urban agglomerations using ensemble learning

Urban extreme heat (UEH) has been imparting a formidable menace to various facets of sustainable urban and regional development. While increasing studies have examined the urban thermal environment, a gap exists in decoupling the nonlinear multi-drivers of UEH environment in intra-urban regions, especially across the urban agglomerations. In this study, we estimated UEH through the land surface temperature (LST) in the extended summer season exceeding a specific threshold computed through cross-regional statistics. Furthermore, an additive interpretable ensemble learning model enhanced via the Bayesian Optimization algorithm (BO) and Monte Carlo Simulation framework (MCS) was employed to dissect the intricate nonlinear interplay between UEH and intra-urban morphology, green-blue composition, and atmospheric determinants. Our results highlight the indispensably suppressive effect of green and blue factors on UEH, whereas urban morphological variables generally exhibit an opposite trend. Interestingly, we reveal heterogeneity responses of building heights, elevated urban structures mitigate UEH in temperate monsoon and subtropical monsoon zones and exhibit diminished marginal utility in temperate continental zones. Furthermore, the arid climatic characteristics are expected to exert an unexpectedly enhancing effect on water bodies' cooling capacity. These findings provide heterogeneous local guidance for landscape and urban planning towards the severe challenges of climate change.

Exploring the Impact of Pandemic Measures on Airport Performance

The impact of COVID-19 measures on airport performance is obvious, and there have been numerous studies on this topic. However, most of these studies discuss prevention measures, the effects on airport operations, forecasts of economic impacts, changes in service quality, etc. There is a lack of research on the effects of various prevention measures on airport operations and the interrelationships between these measures. This study focuses on addressing this gap. In this study, an integrated approach is devised that combines the decision-making trial and evaluation laboratory (DEMATEL) method and interpretive structural modeling (ISM). This integrated method is useful for exploring the relationship between pandemic measures and airport performance as well as the complex relationship between them, and the combination of methods improves upon the shortcomings of the original models. This study reveals that mandating vaccination certificates for entry into a country is the most significant measure affecting airport performance. Additionally, aircraft movement at the airport has the greatest overall impact and can be considered the most crucial factor influencing airport performance from an operational standpoint. The findings show that both factors directly influence financial performance, as reflected in the net income. Some management implications are provided to mitigate the consequences of the measures taken to counter the pandemic crisis. This integrated approach should also assist authorities and policy-makers in planning cautious action for future crises.

Supply chain collaboration, agility and firm performance: a case of manufacturing SMEs in India

PurposeThe dynamic business environment and intense competition have mandated agility in operations for manufacturing firms. Effective inter-organizational collaboration can make operations more agile. This paper develops an interpretive model to explore the effect of supply chain collaboration (SCC) on supply chain agility (SCA) and, subsequently, on business performance.Design/methodology/approachA hierarchical model that illustrates the relationship between SCC, SCA and firm performance components is developed using total interpretative structural modeling (TISM). Also, statistical validation of the model has been performed.FindingsThe results indicate that the vision and alertness of the firm on the strategic front promote collaboration between supply chain partners. This creates operational agility, helping the firm to absorb fluctuations in demand, thereby enhancing business performance.Research limitations/implicationsThe opinion of most respondents was considered to develop the TISM framework over the fuzzy one, which necessitates a significantly more extensive data set. However, multiple discussions with participants can eliminate the prejudice of the majority approach. Also, the paper's development and validation were restricted to Indian manufacturing small and medium-sized enterprises (SMEs). The model can potentially be evaluated in large organizations to provide further insights.Originality/valueThe study blends the factors of SCC and SCA in a novel way to explain their combined impact on business performance. The TISM model addresses the “why” of theory development in addition to the “what” and “how” of it. Using triangulation in combination with the interpretative tool, this study additionally offers methodological value.

Automated detection of steno-occlusive lesion on time-of-flight magnetic resonance angiography: an observer performance study.

Intracranial steno-occlusive lesions are responsible for acute ischemic stroke. However, the clinical benefits of artificial intelligence-based methods for detecting pathologic lesions in intracranial arteries have not been evaluated. We aimed to validate the clinical utility of an artificial intelligence model for detecting steno-occlusive lesions in the intracranial arteries. Overall, 138 TOF-MRA images were collected from two institutions, which served as internal (n = 62) and external (n = 76) test sets, respectively. Each study was reviewed by five radiologists (two neuroradiologists and three radiology residents) to compare the usage and non-usage of our proposed artificial intelligence model for TOF-MRA interpretation. They identified the steno-occlusive lesions and recorded their reading time. Observer performance was assessed using the area under the Jackknife free-response receiver operating characteristic curve and reading time for comparison. The average area under the Jackknife free-response receiver operating characteristic curve for the five radiologists demonstrated an improvement from 0.70 without artificial intelligence to 0.76 with artificial intelligence (P = .027). Notably, this improvement was most pronounced among the three radiology residents, whose performance metrics increased from 0.68 to 0.76 (P = .002). Despite an increased reading time upon using artificial intelligence, there was no significant change among the readings by radiology residents. Moreover, the use of artificial intelligence resulted in improved inter-observer agreement among the reviewers (the intraclass correlation coefficient increased from 0.734 to 0.752). Our proposed artificial intelligence model offers a supportive tool for radiologists, potentially enhancing the accuracy of detecting intracranial steno-occlusion lesions on TOF-MRA. Less-experienced readers may benefit the most from this model.ABBREVIATIONS: AI = Artificial intelligence; AUC = Area under the receiver operating characteristic curve; AUFROC = Area under the Jackknife free-response receiver operating characteristic curve; DL = Deep learning; ICC = Intraclass correlation coefficient; IRB = Institutional Review Boards; JAFROC = Jackknife free-response receiver operating characteristic.

Unraveling the relationship between coastal landscapes and sentiments: An integrated approach based on social media data and interpretable machine learning methods

AbstractCoastal landscapes exert a significant impact on the human sentimental perceptions and physical and mental well‐being of people. However, little is known about explicitly linking between the landscape characteristics and people's sentimental preferences expressed in social media data. The main objective of this study was to explore the nonlinear and interaction effects of key factors that influenced sentiments in the coastal areas of Hong Kong, considering both subjective landscape preferences and objective landscape patterns. We quantified users' sentiment polarity based on the crowdsourcing textual data of Flickr. To study users' subjective landscape preferences, we computed various visual landscape objects' proportion in images. Meanwhile, eight user clusters and nine image clusters were detected by the identified visual object labels. We quantified objective landscape patterns considering the land use pattens and the availability of public service facilities. Finally, we utilized an interpretable classification model to analyze the factors that may affect sentiments and their interplay interactions. We found that ecotourism‐related clusters exhibited the most positive sentiment. The proportion of floor and sky pixels in images exhibits the highest global relative importance when predicting sentiments. This study extends a new insight on the relationship between landscape characteristics and sentiments from both subjective and objective perspectives based on social media data and interpretable machine learning methods. This research may help decision‐makers in designing landscapes that aptly satisfy to the needs of the public and promote sustainable management of the coastal environment.

ROBUST LEARNING AND ACCURATE PREDICTION OF WIRELESS TRAFFIC USING IMPROVED NEURAL NETWORK

The proposed system introduces a novel Deep Metric Algorithm as a cornerstone for enhancing the feasibility and optimality of network traffic prediction within the dynamic landscape of cloud technologies. The Deep Metric Algorithm is designed to operate as a physically interpretable probabilistic model that captures network-wide traffic patterns. Unlike traditional methods that rely solely on current traffic states observed at specific intervals, this algorithm leverages an initially expensive set of measurements to construct a network-specific traffic model.The algorithm's innovation lies in its ability to utilize this network-specific model alongside readily available, less expensive measurements, thereby significantly improving the accuracy of traffic predictions. By employing the Deep Metric Algorithm, the system can predict traffic fluctuations not only over observed links but also extend its predictions to unobserved links. This is a crucial advancement as it enables the system to offer more comprehensive insights into network behavior, even where direct measurements might be limited.One notable strength of the proposed Deep Metric Algorithm is its demonstrated applicability across various traffic periods within the same network. This adaptability ensures that the learned model remains effective over time, showcasing its potential utility in optimizing network performance as it evolves. The algorithm's predictive capabilities, coupled with its ability to adapt to changing network conditions, position it as a valuable tool for load-aware resource management and predictive control, thereby contributing to the efficient management of multi-provider networks in the evolving cloud landscape. KEYWORDS: Quality of Service (QoS), network traffic, accurate predictions, network-specific traffic model, Deep Metric Algorithm.

A physics and data co-driven surrogate modeling method for high-dimensional rare event simulation

This paper presents a physics and data co-driven surrogate modeling method for efficient rare event simulation of civil and mechanical systems with high-dimensional input uncertainties. The method fuses interpretable low-fidelity physical models with data-driven error corrections. The hypothesis is that a well-designed and well-trained simplified physical model can preserve salient features of the original model, while data-fitting techniques can fill the remaining gaps between the surrogate and original model predictions. The coupled physics-data-driven surrogate model is adaptively trained using active learning, aiming to achieve a high correlation and small bias between the surrogate and original model responses in the critical parametric region of a rare event. A final importance sampling step is introduced to correct the surrogate model-based probability estimations. Static and dynamic problems with input uncertainties modeled by random field and stochastic process are studied to demonstrate the proposed method.

Developing interpretable machine learning model for evaluating young modulus of cemented paste backfill

Cemented paste backfill (CPB), a mixture of wet tailings, binding agent, and water, proves cost-effective and environmentally beneficial. Determining the Young modulus during CPB mix design is crucial. Utilizing machine learning (ML) tools for Young modulus evaluation and prediction streamlines the CPB mix design process. This study employed six ML models, including three shallow models Extreme Gradient Boosting (XGB), Gradient Boosting (GB), Random Forest (RF) and three hybrids Extreme Gradient Boosting-Particle Swarm Optimization (XGB-PSO), Gradient Boosting-Particle Swarm Optimization (GB-PSO), Random Forest-Particle Swarm Optimization (RF-PSO). The XGB-PSO hybrid model exhibited superior performance (coefficient of determination R2 = 0.906, root mean square error RMSE = 19.535 MPa, mean absolute error MAE = 13.741 MPa) on the testing dataset. Shapley Additive Explanation (SHAP) values and Partial Dependence Plots (PDP) provided insights into component influences. Cement/Tailings ratio emerged as the most crucial factor for enhancing Young modulus in CPB. Global interpretation using SHAP values identified six essential input variables: Cement/Tailings, Curing age, Cc, solid content, Fe2O3 content, and SiO2 content.

DIRECT: Dual Interpretable Recommendation with Multi-aspect Word Attribution

Recommending products to users with intuitive explanations helps improve the system in transparency, persuasiveness, and satisfaction. Existing interpretation techniques include post-hoc methods and interpretable modeling. The former category could quantitatively analyze input contribution to model prediction but has limited interpretation faithfulness, while the latter could explain model internal mechanisms but may not directly attribute model predictions to input features. In this study, we propose a novel D ual I nterpretable Rec ommenda t ion model called DIRECT, which integrates ideas of the two interpretation categories to inherit their advantages and avoid limitations. Specifically, DIRECT makes use of item descriptions as explainable evidence for recommendation. First, similar to the post-hoc interpretation, DIRECT could attribute the prediction of a user preference score to textual words of the item descriptions. The attribution of each word is related to its sentiment polarity and word importance, where a word is important if it corresponds to an item aspect that the user is interested in. Second, to improve the interpretability of embedding space, we propose to extract high-level concepts from embeddings, where each concept corresponds to an item aspect. To learn discriminative concepts, we employ a concept-bottleneck layer, and maximize the coding rate reduction on word-aspect embeddings by leveraging a word-word affinity graph extracted from a pre-trained language model. In this way, DIRECT simultaneously achieves faithful attribution and usable interpretation of embedding space. We also show that DIRECT achieves linear inference time complexity regarding the length of item reviews. We conduct experiments including ablation studies on five real-world datasets. Quantitative analysis, visualizations, and case studies verify the interpretability of DIRECT. Our code is available at: https://github.com/JacksonWuxs/DIRECT .

The influence of corporate social responsibility on impulse buying

PurposeCorporate social responsibility (CSR) ISO standards have been noted as an essential marketing strategy by which firms can achieve consumer trust while improving environmental, social, and quality factors. This study discloses the contextual relationship between CSR ISO standards and sustainable impulse buying behavior. This study also looks to uncover the CSR ISO driving and linkage factors that motivate consumers to make sustainable impulsive purchases.Design/methodology/approachThree distinct research methods were employed in this research. First, a consumer expert opinion-based Interpretive Structural Modeling (ISM) approach was adopted to reveal the contextual relationship between CSR ISO factors and sustainable impulse buying behavior. Secondly, Matrice Impacts Croises Multiplication Appliques Classement (MICMAC) was used to examine these factors' driving and dependent power. In addition, Minitab package software was also used to check the statistical validation of ISM-MICMAC results.FindingsThe results indicate that although environmentally responsible CSR ISO 14001, socially responsible CSR ISO 26000, and consumer perception of product quality CSR ISO 9001 standards contain strong driving power, their dependent power was weak. All these CSR ISO factors (14,001, 26,000, and 9001) strongly impact each other and sustainable impulse buying. Therefore, these three CSR ISO factors have been placed at the bottom of the ISM model. The CSR ISO 14020 standard (labeling of the product), knowledge of CSR ISO standards, consumer trust, and advertising about CSR ISO standards have been placed in the middle. The mentioned factors have intense driving and dependent power and are classified as linkage factors for sustainable impulse buying. Impulse buying behavior has weak driving and strong dependent power, yet this factor strongly depends on other CSR ISO factors. Hence, this factor is placed at the top of the ISM model. In addition, the Minitab package software results indicate that ISM-MICMAC results are statistically valid.Originality/valueTo the best of our knowledge, this research is unique and examines the influence of CSR ISO factors on sustainable impulse buying in the context of Pakistani consumers. Secondly, our study has thoroughly investigated several CSR ISO factors and allied these factors in the context of consumer buying behavior. Third, several CSR ISO factors and impulse buying behavior were examined using a mix of ISM-MICAC and Minitab methods. Thus, including these steps in our study has led to the development of a novel technique.

Pseudo-class part prototype networks for interpretable breast cancer classification

Interpretability in machine learning has become increasingly important as machine learning is being used in more and more applications, including those with high-stakes consequences such as healthcare where Interpretability has been regarded as a key to the successful adoption of machine learning models. However, using confounding/irrelevant information in making predictions by deep learning models, even the interpretable ones, poses critical challenges to their clinical acceptance. That has recently drawn researchers’ attention to issues beyond the mere interpretation of deep learning models. In this paper, we first investigate application of an inherently interpretable prototype-based architecture, known as ProtoPNet, for breast cancer classification in digital pathology and highlight its shortcomings in this application. Then, we propose a new method that uses more medically relevant information and makes more accurate and interpretable predictions. Our method leverages the clustering concept and implicitly increases the number of classes in the training dataset. The proposed method learns more relevant prototypes without any pixel-level annotated data. To have a more holistic assessment, in addition to classification accuracy, we define a new metric for assessing the degree of interpretability based on the comments of a group of skilled pathologists. Experimental results on the BreakHis dataset show that the proposed method effectively improves the classification accuracy and interpretability by respectively 8%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$8 \\%$$\\end{document} and 18%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$18 \\%$$\\end{document}. Therefore, the proposed method can be seen as a step toward implementing interpretable deep learning models for the detection of breast cancer using histopathology images.

Interpretable and explainable hybrid model for daily streamflow prediction based on multi-factor drivers.

Streamflow time series data typically exhibit nonlinear and nonstationary characteristics that complicate precise estimation. Recently, multifactorial machine learning (ML) models have been developed to enhance the performance of streamflow predictions. However, the lack of interpretability within these ML models raises concerns about their inner workings and reliability. This paper introduces an innovative hybrid architecture, the TCN-LSTM-Multihead-Attention model, which combines two layers of temporal convolutional networks (TCN) followed by one layer of long short-term memory (LSTM) units, integrated with a Multihead-Attention mechanism for predicting streamflow with streamflow causation-driven prediction samples (RCDP), employing local and global interpretability studies through Shapley values and partial dependency analysis. The find_peaks method was used to identify peak flow events in the test dataset, validating the model's generality and uncovering the physical causative patterns of streamflow. The results show that (1) compared to the LSTM model with the same hyperparameter settings, the proposed TCN-LSTM-Multihead-Attention hybrid model increased the R2 by 52.9%, 2.5%, 43.1%, and 10.7% respectively at four stations in the test set predictions using RCDP samples. Moreover, comparing the prediction results of the hybrid model under different samples in Hengshan station, the R2 for RCDP increased by 5.06% and 1.22% compared to streamflow autoregressive prediction samples (RAP) and meteorological-soil volumetric water content coupled autoregressive prediction samples (MCSAP) respectively. (2) Historical streamflow data from the preceding 3days predominantly influences predictions due to strong autocorrelation, with flow quantity (Q) typically emerging as the most significant feature alongside precipitation (P), surface soil moisture (SSM), and adjacent station flow data. (3) During periods of low and normal flow, historical data remains the most crucial factor; however, during flood periods, the roles of upstream inflow and precipitation become significantly more pronounced. This model facilitates the identification and quantification of various hydrodynamic impacts on flow predictions, including upstream flood propagation, precipitation, and soil moisture conditions. It also elucidates the model's nonlinear relationships and threshold responses, thereby enhancing the interpretability and reliability of streamflow predictions.

Interpretable machine learning model for shear strength estimation of circular concrete‐filled steel tubes

SummaryPrecise estimation of the shear strength of concrete‐filled steel tubes (CFSTs) is a crucial requirement for the design of these members. The existing design codes and empirical equations are inconsistent in predicting the shear strength of these members. This paper provides a data‐driven approach for the shear strength estimation of circular CFSTs. For this purpose, the authors evaluated and compared the performance of nine machine learning (ML) methods, namely linear regression, decision tree (DT), k‐nearest neighbors (KNN), support vector regression (SVR), random forest (RF), bagging regression (BR), adaptive boosting (AdaBoost), gradient boosting regression tree (GBRT), and extreme gradient boosting (XGBoost) in estimating the shear strength of CFSTs on an experimental database compiled from the results of 230 shear tests on CFSTs in the literature. For each model, hyperparameter tuning was performed by conducting a grid search in combination with k‐fold cross‐validation (CV). Comparing the nine methods in terms of several performance measures showed that the XGBoost model was the most accurate in predicting the shear strength of CFSTs. This model also showed superior accuracy in predicting the shear strength of CFSTs when compared to the formulas provided in design codes and the existing empirical equations. The Shapley Additive exPlanations (SHAP) technique was also used to interpret the results of the XGBoost model. Using SHAP, the features with the greatest impact on the shear strength of CFSTs were found to be the cross‐sectional area of the steel tube, the axial load ratio, and the shear span ratio, in that order.

An aquaculture risk model to understand the causes and consequences of Atlantic Salmon mass mortality events: A review

AbstractMass mortality events (MMEs) are defined as the death of large numbers of fish over a short period of time. These events can result in catastrophic losses to the Atlantic salmon aquaculture industry and the local economy. However, they are challenging to understand because of their relative infrequency and the high number of potential factors involved. As a result, the causes and consequences of MMEs in Atlantic salmon aquaculture are not well understood. In this study, we developed a structural network of causal risk factors for MMEs for aquaculture and the communities that depend on Atlantic salmon aquaculture. Using the Interpretive Structural Modeling (ISM) technique, we analysed the causes of Atlantic salmon mass mortalities due to environmental (abiotic), biological (biotic) and nutritional risk factors. The consequences of MMEs were also assessed for the occupational health and safety of aquaculture workers and their implications for the livelihoods of local communities. This structural network deepens our understanding of MMEs and points to management actions and interventions that can help mitigate mass mortalities. MMEs are typically not the result of a single risk factor but are caused by the systematic interaction of risk factors related to the environment, fish diseases, feeding/nutrition and cage‐site management. Results also indicate that considerations of health and safety risk, through pre‐ and post‐event risk assessments, may help to minimize workplace injuries and eliminate potential risks of human fatalities. Company and government‐assisted socio‐economic measures could help mitigate post‐mass mortality impacts. Appropriate and timely management actions may help reduce MMEs at Atlantic salmon cage sites and minimize the physical and social vulnerabilities of workers and local communities.