Feature Importance Measure Research Articles

A wrapper feature selection approach using Markov blankets

In feature selection, Markov Blanket (MB) based approaches have attracted considerable attention with most MB discovery algorithms being categorized as filter based techniques. Typically, the Conditional Independence (CI) test employed by such methods is different for different data types. In this article, we propose a novel Markov Blanket based wrapper feature selection method. The proposed approach employs Predictive Permutation Independence (PPI), a novel Conditional Independence (CI) test that allows it to work out-of-the-box for both classification and regression tasks on mixed data. PPI can work with any supervised algorithm to estimate the association of a feature with the target variable while also providing a measure of feature importance. The proposed approach also includes an optional MB aggregation step that can be used to find the optimal MB under non-faithful conditions. Our method11Implementation and experimental results are available at https://github.com/AnonymousMLSubmissions/PatternRecognitionSubmission. outperforms other MB discovery methods, in terms of F1-score, by 7% on average, over 3 large-scale BN datasets. It also outperforms state-of-the-art feature selection techniques on 13 real-world datasets.

Multi-Feature Extraction and Selection Method to Diagnose Burn Depth from Burn Images

Burn wound depth is a significant determinant of patient treatment. Typically, the evaluation of burn depth relies heavily on the clinical experience of doctors. Even experienced surgeons may not achieve high accuracy and speed in diagnosing burn depth. Thus, intelligent burn depth classification is useful and valuable. Here, an intelligent classification method for burn depth based on machine learning techniques is proposed. In particular, this method involves extracting color, texture, and depth features from images, and sequentially cascading these features. Then, an iterative selection method based on random forest feature importance measure is applied. The selected features are input into the random forest classifier to evaluate this proposed method using the standard burn dataset. This method classifies burn images, achieving an accuracy of 91.76% when classified into two categories and 80.74% when classified into three categories. The comprehensive experimental results indicate that this proposed method is capable of learning effective features from limited data samples and identifying burn depth effectively.

Interpretation techniques to explain the output of a spatial land subsidence hazard model in an area with a diverted tributary

Due to the nature of black-box machine learning (ML) models used in the spatial modelling field of environmental and natural hazards, the interpretation of predictive model outputs is necessary. For this purpose, we applied four interpretation techniques consisting of interaction plot, permutation feature importance (PFI) measure, shapley additive explanation (SHAP) decision plot, and accumulated local effects (ALE) plot to explain and interpret the output of an ML model applied to map land subsidence (LS) in the Nazdasht plain, Hormozgan province, southern Iran. We applied a stepwise regression (SR) algorithm and five ML models (Cforest (as a conditional random forest), generalized linear model (GLM), multivariate linear regression (MLR), partial least squares (PLS) and extreme gradient boosting (XGBoost)) to select important features and to map the LS hazard, respectively. Thereafter, several interpretation techniques were used to explain the spatial ML hazard model output. Our findings revealed that a GLM model was the most accurate approach to map LS in our study area, and that 24.3% of the total study area had a very high susceptibility to the LS hazard. According to the interpretation techniques, land use, elevation, groundwater level and vegetation were the most important variables controlling the LS hazard and also the most important variables contributing to the model’s output. Overall, human activities, especially the diversion of the route of one of the main tributaries feeding the plain and the recharging of groundwater five decades ago, intensified the current LS occurrence. Therefore, management activities such as water spreading projects upstream of the plain can be useful to mitigate LS occurrence in the plain.

COPDVD: Automated classification of chronic obstructive pulmonary disease on a new collected and evaluated voice dataset

BackgroundChronic obstructive pulmonary disease (COPD) is a severe condition affecting millions worldwide, leading to numerous annual deaths. The absence of significant symptoms in its early stages promotes high underdiagnosis rates for the affected people. Besides pulmonary function failure, another harmful problem of COPD is the systemic effects, e.g., heart failure or voice distortion. However, the systemic effects of COPD might provide valuable information for early detection. In other words, symptoms caused by systemic effects could be helpful to detect the condition in its early stages. ObjectiveThe proposed study aims to explore whether the voice features extracted from the vowel “a” utterance carry any information that can be predictive of COPD by employing Machine Learning (ML) on a newly collected voice dataset. MethodsForty-eight participants were recruited from the pool of research clinic visitors at Blekinge Institute of Technology (BTH) in Sweden between January 2022 and May 2023. A dataset consisting of 1246 recordings from 48 participants was gathered. The collection of voice recordings containing the vowel “a” utterance commenced following an information and consent meeting with each participant using the VoiceDiagnostic application. The collected voice data was subjected to silence segment removal, feature extraction of baseline acoustic features, and Mel Frequency Cepstrum Coefficients (MFCC). Sociodemographic data was also collected from the participants. Three ML models were investigated for the binary classification of COPD and healthy controls: Random Forest (RF), Support Vector Machine (SVM), and CatBoost (CB). A nested k-fold cross-validation approach was employed. Additionally, the hyperparameters were optimized using grid-search on each ML model. For best performance assessment, accuracy, F1-score, precision, and recall metrics were computed. Afterward, we further examined the best classifier by utilizing the Area Under the Curve (AUC), Average Precision (AP), and SHapley Additive exPlanations (SHAP) feature-importance measures. ResultsThe classifiers RF, SVM, and CB achieved a maximum accuracy of 77 %, 69 %, and 78 % on the test set and 93 %, 78 % and 97 % on the validation set, respectively. The CB classifier outperformed RF and SVM. After further investigation of the best-performing classifier, CB demonstrated the highest performance, producing an AUC of 82 % and AP of 76 %. In addition to age and gender, the mean values of baseline acoustic and MFCC features demonstrate high importance and deterministic characteristics for classification performance in both test and validation sets, though in varied order. ConclusionThis study concludes that the utterance of vowel “a” recordings contain information that can be captured by the CatBoost classifier with high accuracy for the classification of COPD. Additionally, baseline acoustic and MFCC features, in conjunction with age and gender information, can be employed for classification purposes and benefit healthcare for decision support in COPD diagnosis. Clinical trial registration numberNCT05897944.

Dam Deformation Prediction Considering the Seasonal Fluctuations Using Ensemble Learning Algorithm

Dam deformation is the most visual and relevant monitoring quantity that reflects the operational condition of a concrete dam. The seasonal variations in the external environment can induce seasonal fluctuations in the deformation of concrete dams. Hence, preprocessing the deformation monitoring series to identify seasonal fluctuations within the series can effectively enhance the accuracy of the predictive model. Firstly, the dam deformation time series are decomposed into the seasonal and non-seasonal components based on the seasonal decomposition technique. The advanced ensemble learning algorithm (Extreme Gradient Boosting model) is used to forecast the seasonal and non-seasonal components independently, as well as employing the Tree-structured Parzen Estimator (TPE) optimization algorithm to tune the model parameters, ensuring the optimal performance of the prediction model. The results of the case study indicate that the predictive performance of the proposed model is intuitively superior to the benchmark models, demonstrated by a higher fitting accuracy and smaller prediction residuals. In the comparison of the objective evaluation metrics RMSE, MAE, and R2, the proposed model outperforms the benchmark models. Additionally, using feature importance measures, it is found that in predicting the seasonal component, the importance of the temperature component increases, while the importance of the water pressure component decreases compared to the prediction of the non-seasonal component. The proposed model, with its elevated predictive accuracy and interpretability, enhances the practicality of the model, offering an effective approach for predicting concrete dam deformation.

Random Forest and Feature Importance Measures for Discriminating the Most Influential Environmental Factors in Predicting Cardiovascular and Respiratory Diseases.

Several studies suggest that environmental and climatic factors are linked to the risk of mortality due to cardiovascular and respiratory diseases; however, it is still unclear which are the most influential ones. This study sheds light on the potentiality of a data-driven statistical approach by providing a case study analysis. Daily admissions to the emergency room for cardiovascular and respiratory diseases are jointly analyzed with daily environmental and climatic parameter values (temperature, atmospheric pressure, relative humidity, carbon monoxide, ozone, particulate matter, and nitrogen dioxide). The Random Forest (RF) model and feature importance measure (FMI) techniques (permutation feature importance (PFI), Shapley Additive exPlanations (SHAP) feature importance, and the derivative-based importance measure (κALE)) are applied for discriminating the role of each environmental and climatic parameter. Data are pre-processed to remove trend and seasonal behavior using the Seasonal Trend Decomposition (STL) method and preliminary analyzed to avoid redundancy of information. The RF performance is encouraging, being able to predict cardiovascular and respiratory disease admissions with a mean absolute relative error of 0.04 and 0.05 cases per day, respectively. Feature importance measures discriminate parameter behaviors providing importance rankings. Indeed, only three parameters (temperature, atmospheric pressure, and carbon monoxide) were responsible for most of the total prediction accuracy. Data-driven and statistical tools, like the feature importance measure, are promising for discriminating the role of environmental and climatic factors in predicting the risk related to cardiovascular and respiratory diseases. Our results reveal the potential of employing these tools in public health policy applications for the development of early warning systems that address health risks associated with climate change, and improving disease prevention strategies.

Poster 387: Application of Survival Machine Learning Methodology to Predict Time-Dependent Risk of Revision ACLR Graft Failure in the MARS Cohort

Objectives: Previous machine learning (ML) analysis has shown that a novel ML model built using the Multicenter Anterior Cruciate Ligament (ACL) Revision Study (MARS) cohort data can predict graft failure at 6 years postoperatively as a binary outcome with moderate ability. However, binary classification does not consider the time-to-event component of graft failure as an outcome. In contrast, survival models can factor the time-to-event component of an outcome, incorporate time-dependent variables, and appropriately process censored data to deliver more clinically meaningful outcome predictions. While the previous binary classifier can provide information about whether the primary outcome graft failure will occur at 6 years, survival analysis models are able to provide information on when graft failure occurs (for example, within 2, 4, or 6 years). Therefore, this study sought to apply ML survival analysis methodology to the MARS cohort data to produce a model that can most accurately estimate the probability of revision ACL reconstruction (rACLR) graft failure within 2, 4, and 6 years postoperatively, and identify factors that impact probability estimates at each of these time horizons. Methods: rACLR patients were prospectively recruited by the MARS group. Preoperative radiographs, surgeon-reported intraoperative findings, and 2 and 6-year follow-up data on patient-reported outcomes (PROs), subsequent knee surgeries, and graft failure were obtained. ML models including Cox Proportional Hazards (CoxPH), XGBoost Survival Embeddings (XGBoost), Random Survival Forest (RSF), Extra Survival Tree (EST), and CoxBoost, were built to predict the probability of graft failure within 2, 4, and 6 years postoperatively. Validated performance metrics and feature importance measures were used to evaluate model performance. Results: The cohort included 1142 patients, of which 4.8% (n = 55) experienced documented graft failure during the 6-year postoperative period. CoxBoost demonstrated the highest discriminative power across all studied time points (Timepoint: Time Dependent C-index | 2-year: 0.725 ± 0.127 | 4-year: 0.669 ± 0.057 | 6-year: 0.645 ± 0.054), with well-calibrated scores (Timepoint: Time dependent Brier Score | 2-year: 0.023 ± 0.012 | 4-year: 0.035 ± 0.010 | 6-year: 0.046 ± 0.012) as listed in Table 1 and Table 2 respectively. Based on the Kaplan Meier survival plot, the survival probability (probability of not tearing the ACL graft) at 2, 4, 6 years was 0.977, 0.965, and 0.954 respectively. Within the context of the predictive models, features deemed important for CoxBoost differed from those deemed important for CoxPH predictive ability. Conclusions: Survival machine learning models can predict the risk of rACLR graft failure up to 6 years postoperatively, with the most robust predictions at the 2-year postoperative timepoint. While external validation with further combined registry datasets is required, this present study builds on prior ML analyses towards the development of a bedside calculator for rACLR outcome prediction and risk stratification.

The many Shapley values for explainable artificial intelligence: A sensitivity analysis perspective

Predictive models are increasingly used for managerial and operational decision-making. The use of complex machine learning algorithms, the growth in computing power, and the increase in data acquisitions have amplified the black-box effects in data science. Consequently, a growing body of literature is investigating methods for interpretability and explainability. We focus on methods based on Shapley values, which are gaining attention as measures of feature importance for explaining black-box predictions. Our analysis follows a hierarchy of value functions, and proves several theoretical properties that connect the indices at the alternative levels. We bridge the notions of totally monotone games and Shapley values, and introduce new interaction indices based on the Shapley-Owen values. The hierarchy evidences synergies that emerge when combining Shapley effects computed at different levels. We then propose a novel sensitivity analysis setting that combines the benefits of both local and global Shapley explanations, which we refer to as the “glocal” approach. We illustrate our integrated approach and discuss the managerial insights it provides in the context of a data-science problem related to health insurance policy-making.

Sensitivity and feature importance of climate factors and evaluation of different machine learning models for predicting fire hotspots in Kalimantan, Indonesia

<p>Hotspots as indicators of forest fires capable of quickly monitoring large areas are often predicted using various machine learning methods. However, there is still little research that analyzes the sensitivity and feature importance of each predictor that forms a machine learning prediction model. This study evaluates and compares several machine learning methods to predict hotspots in Kalimantan. Using the most accurate machine learning model, each climate factor used as a predictor is analyzed for its sensitivity and feature importance. Some of the machine learning methods used include random forest, gradient boosting, Bayesian regression, and artificial neural networks. Meanwhile, several measures of sensitivity and feature importance used are variance-based, density-based, and distribution-based sensitivity indices, as well as permutation and Shapley feature importance. Evaluation of the ML model concluded that the Bayesian linear regression model outperformed other ML models, based on RMSE and explained variance score. Meanwhile, tree-based models, such as random forest and gradient boosting, are indicative of overfit. Based on the results of sensitivity analysis and feature importance, the number of dry days is the most important feature for the Bayesian linear regression model in predicting the number of hotspots in Kalimantan.</p>

A game theoretic decision forest for feature selection and classification

Abstract Classification and feature selection are two of the most intertwined problems in machine learning. Decision trees (DTs) are straightforward models that address these problems offering also the advantage of explainability. However, solutions that are based on them are either tailored for the problem they solve or their performance is dependent on the split criterion used. A game-theoretic decision forest model is proposed to approach both issues. DTs in the forest use a splitting mechanism based on the Nash equilibrium concept. A feature importance measure is computed after each tree is built. The selection of features for the next trees is based on the information provided by this measure. To make predictions, training data is aggregated from all leaves that contain the data tested, and logistic regression is further used. Numerical experiments illustrate the efficiency of the approach. A real data example that studies country income groups and world development indicators using the proposed approach is presented.

On relevant features for the recurrence prediction of urothelial carcinoma of the bladder

BackgroundUrothelial bladder cancer (UBC) is characterized by a high recurrence rate, which is predicted by scoring systems. However, recent studies show the superiority of Machine Learning (ML) models. Nevertheless, these ML approaches are rarely used in medical practice because most of them are black-box models, that cannot adequately explain how a prediction is made. ObjectiveWe investigate the global feature importance of different ML models. By providing information on the most relevant features, we can facilitate the use of ML in everyday medical practice. Design, setting, and participantsThe data is provided by the cancer registry Rhineland-Palatinate gGmbH, Germany. It consists of numerical and categorical features of 1,944 patients with UBC. We retrospectively predict 2-year recurrence through ML models using Support Vector Machine, Gradient Boosting, and Artificial Neural Network. We then determine the global feature importance using performance-based Permutation Feature Importance (PFI) and variance-based Feature Importance Ranking Measure (FIRM). ResultsWe show reliable recurrence prediction of UBC with 82.02% to 83.89% F1-Score, 83.95% to 84.49% Precision, and an overall performance of 69.20% to 70.82% AUC on testing data, depending on the model. Gradient Boosting performs best among all black-box models with an average F1-Score (83.89%), AUC (70.82%), and Precision (83.95%). Furthermore, we show consistency across PFI and FIRM by identifying the same features as relevant across the different models. These features are exclusively therapeutic measures and are consistent with findings from both medical research and clinical trials. ConclusionsWe confirm the superiority of ML black-box models in predicting UBC recurrence compared to more traditional logistic regression. In addition, we present an approach that increases the explanatory power of black-box models by identifying the underlying influence of input features, thus facilitating the use of ML in clinical practice and therefore providing improved recurrence prediction through the application of black-box models.

Feature importance measures for flood forecasting system design

ABSTRACT Effective proxy selection in hydrological processes is crucial in several applications. This study investigates the role of sub-basins in hydrological response, which remains unclear. Our focus is on exploring feature importance measures to identify influential sub-basins in a flood forecasting system. We use the Tiber River basin as a case study and employ a synthetic flood hydrograph dataset, consisting in approximately 20 000 simulated annual maximum hydrographs across 39 sub-basins and the basin outlet. Through this study, we present a proof of concept for ranking sub-basins based on their contribution to basin response using six feature importance measures. The results reveal eight influential sub-basins and provide guidance for strategically installing measurement instrumentation for an efficient and cost-effective flood early warning system.

Total effects with constrained features

Recent studies have emphasized the connection between machine learning feature importance measures and total order sensitivity indices (total effects, henceforth). Feature correlations and the need to avoid unrestricted permutations make the estimation of these indices challenging. Additionally, there is no established theory or approach for non-Cartesian domains. We propose four alternative strategies for computing total effects that account for both dependent and constrained features. Our first approach involves a generalized winding stairs design combined with the Knothe-Rosenblatt transformation. This approach, while applicable to a wide family of input dependencies, becomes impractical when inputs are physically constrained. Our second approach is a U-statistic that combines the Jansen estimator with a weighting factor. The U-statistic framework allows the derivation of a central limit theorem for this estimator. However, this design is computationally intensive. Then, our third approach uses derangements to significantly reduce computational burden. We prove consistency and central limit theorems for these estimators as well. Our fourth approach is based on a nearest-neighbour intuition and it further reduces computational burden. We test these estimators through a series of increasingly complex computational experiments with features constrained on compact and connected domains (circle, simplex), non-compact and non-connected domains (Sierpinski gaskets), we provide comparisons with machine learning approaches and conclude with an application to a realistic simulator.

The most efficient position of magnets

People mount materials ranging from paper to wedding photo frames, including school notices to the refrigerators every day at home. The position of the mounts used is essential for both safety and durability. We investigated the most efficient way of positioning magnets that would hold the most pieces of paper on the surface of a refrigerator. We estimated our model coefficients using a regression model with the help of an artificial neural network and decision tree and focused specially on the feature importance measure. We found that when the four magnets are placed in symmetry while the rectangle formed (by drawing a line between the four magnets so that the four magnets become the vertices of the rectangle) contained the center of gravity of the papers, the magnets held more paper. The most efficient points of four magnets predicted in our analysis are in between the center of gravity of the papers and four centers of gravities of four quadrants of a paper. Our study not only considers conditions of efficiency, but also safety, aesthetics, and facility of comprehension. Our findings provide insight into magnet efficiency as resources are scarce, safe for applications, aesthetically pleasing, and easy for individuals to comprehend.

Significance of handcrafted features in human activity recognition with attention-based RNN models

Sensors incorporated in devices are a source of temporal data that can be interpreted to learn the context of a user. The smartphone accelerometer sensor generates data streams that form distinct patterns in response to user activities. The human context can be predicted using deep learning models built from raw sensor data or features retrieved from raw data. This study analyzes data streams from the UCI-HAR public dataset for activity recognition to determine 31 handcrafted features in the temporal and frequency domain. Various stacked and combination RNN models, trained with attention mechanisms, are designed to work with computed features. Attention gave the models a good fit. When trained with all features, the two-stacked GRU model performed best with 99% accuracy. Selecting the most promising features helps reduce training time without compromising accuracy. The ranking supplied by the permutation feature importance measure and Shapley values are utilized to identify the best features from the highly correlated features. Models trained using optimal features, as determined by the importance measures, had a 96% accuracy rate. Misclassification in attention-based classifiers occurs in the prediction of dynamic activities, such as walking upstairs and walking downstairs, and in sedentary activities, such as sitting and standing, due to the similar range of each activity’s axis values. Our research emphasizes the design of streamlined neural network architectures, characterized by fewer layers and a reduced number of neurons when compared to existing models in the field, to design lightweight models to be implemented in resource-constraint gadgets.

Intrinsic and extrinsic techniques for quantification uncertainty of an interpretable GRU deep learning model used to predict atmospheric total suspended particulates (TSP) in Zabol, Iran during the dusty period of 120-days wind

Total suspended particulates (TSP), as a key pollutant, is a serious threat for air quality, climate, ecosystems and human health. Therefore, measurements, prediction and forecasting of TSP concentrations are necessary to mitigate their negative effects. This study applies the gated recurrent unit (GRU) deep learning model to predict TSP concentrations in Zabol, Iran, during the dust period of the 120-day wind (3 June - 4 October 2014). Three uncertainty quantification (UQ) techniques consisting of the blackbox metamodel, heteroscedastic regression and infinitesimal jackknife were applied to quantify the uncertainty associated with GRU model. Permutation feature importance measure (PFIM), based on the game theory, was employed for the interpretability of the predictive model's outputs. A total of 80 TSP samples were collected and were randomly divided as training (70%) and validation (30%) datasets, while eight variables were used in the TSP prediction model. Our findings showed that GRU performed very well for TSP prediction (with r and Nash Sutcliffe coefficient (NSC) values above 0.99 for both datasets, and RMSE of 57 μg m−3 and 73 μg m−3 for training and validation datasets, respectively). Among the three UQ techniques, the infinitesimal jackknife was the most accurate one, while all the observed and predicted TSP values fell within the continence limitation estimated by the model. PFIM plots showed that wind speed and air humidity were the most and least important variables, respectively, impacting the predictive model's outputs. This is the first attempt of using an interpretable DL model for TSP prediction modelling, recommending that future research should involve aspects of uncertainty and interpretability of the predictive models. Overall, UQ and interpretability techniques have a key role in reducing the impact of uncertainties during optimization and decision making, resulting in better understanding of sophisticated mechanisms related to the predictive model.

Evaluating the Impact of Data Transformation Techniques on the Performance and Interpretability of Software Defect Prediction Models

The performance of software defect prediction (SDP) models determines the priority of test resource allocation. Researchers also use interpretability techniques to gain empirical knowledge about software quality from SDP models. However, SDP methods designed in the past research rarely consider the impact of data transformation methods, simple but commonly used preprocessing techniques, on the performance and interpretability of SDP models. Therefore, in this paper, we investigate the impact of three data transformation methods (Log, Minmax, and Z-score) on the performance and interpretability of SDP models. Through empirical research on (i) six classification techniques (random forest, decision tree, logistic regression, Naive Bayes, K-nearest neighbors, and multilayer perceptron), (ii) six performance evaluation indicators (Accuracy, Precision, Recall, F1, MCC, and AUC), (iii) two interpretable methods (permutation and SHAP), (iv) two feature importance measures (Top-k feature rank overlap and difference), and (v) three datasets (Promise, Relink, and AEEEM), our results show that the data transformation methods can significantly improve the performance of the SDP models and greatly affect the variation of the most important features. Specifically, the impact of data transformation methods on the performance and interpretability of SDP models depends on the classification techniques and evaluation indicators. We observe that log transformation improves NB model performance by 7%–61% on the other five indicators with a 5% drop in Precision. Minmax and Z-score transformation improves NB model performance by 2%–9% across all indicators. However, all three transformation methods lead to substantial changes in the Top-5 important feature ranks, with differences exceeding 2 in 40%–80% of cases (detailed results available in the main content). Based on our findings, we recommend that (1) considering the impact of data transformation methods on model performance and interpretability when designing SDP approaches as transformations can improve model accuracy, and potentially obscure important features, which lead to challenges in interpretation, (2) conducting comparative experiments with and without the transformations to validate the effectiveness of proposed methods which are designed to improve the prediction performance, and (3) tracking changes in the most important features before and after applying data transformation methods to ensure precise and traceable interpretability conclusions to gain insights. Our study reminds researchers and practitioners of the need for comprehensive considerations even when using other similar simple data processing methods.

Land Cover Classification Based on Airborne Lidar Point Cloud with Possibility Method and Multi-Classifier.

As important geospatial data, point cloud collected from an aerial laser scanner (ALS) provides three-dimensional (3D) information for the study of the distribution of typical urban land cover, which is critical in the construction of a "digital city". However, existing point cloud classification methods usually use a single machine learning classifier that experiences uncertainty in making decisions for fuzzy samples in confusing areas. This limits the improvement of classification accuracy. To take full advantage of different classifiers and reduce uncertainty, we propose a classification method based on possibility theory and multi-classifier fusion. Firstly, the feature importance measure was performed by the XGBoost algorithm to construct a feature space, and two commonly used support vector machines (SVMs) were the chosen base classifiers. Then, classification results from the two base classifiers were quantitatively evaluated to define the confusing areas in classification. Finally, the confidence degree of each classifier for different categories was calculated by the confusion matrix and normalized to obtain the weights. Then, we synthesize different classifiers based on possibility theory to achieve more accurate classification in the confusion areas. DALES datasets were utilized to assess the proposed method. The results reveal that the proposed method can significantly improve classification accuracy in confusing areas.

Research on the influence of social network on young people’s capacity of innovation and entrepreneurship based on stochastic forest model

Abstract In this paper, firstly, the social network theory and features are studied in detail, and the key to improving the model prediction performance using the Bagging algorithm is to use the CART regression tree algorithm to make the variance as small as possible while ensuring that the average performance of the model does not drop significantly. Next, the random forest model was improved by using the feature importance measure algorithm and the F-measure weighting algorithm, which improved the overall performance of the random forest model. Then the elements of innovation and entrepreneurship ability of college students were identified, and the basic characteristics of the research subjects were obtained using the scale, and the influence of social networks on innovation and entrepreneurship ability based on the random forest model was studied empirically. The results show that the coefficients of the two dimensions of association strength and heterogeneity in the dimensions of social networks of college students are larger, 0.251 and 0.222, respectively, indicating that it is the network association strength and heterogeneity that have a greater impact on the improvement of innovative thinking ability. This study promotes innovation and entrepreneurship education in colleges and universities, which is important for the improvement of students’ innovation and entrepreneurship ability.

Development and validation of a case definition for problematic menopause in primary care electronic medical records

BackgroundMenopause is a normal transition in a woman’s life. For some women, it is a stage without significant difficulties; for others, menopause symptoms can severely affect their quality of life. This study developed and validated a case definition for problematic menopause using Canadian primary care electronic medical records, which is an essential step in examining the condition and improving quality of care.MethodsWe used data from the Canadian Primary Care Sentinel Surveillance Network including billing and diagnostic codes, diagnostic free-text, problem list entries, medications, and referrals. These data formed the basis of an expert-reviewed reference standard data set and contained the features that were used to train a machine learning model based on classification and regression trees. An ad hoc feature importance measure coupled with recursive feature elimination and clustering were applied to reduce our initial 86,000 element feature set to a few tens of the most relevant features in the data, while class balancing was accomplished with random under- and over-sampling. The final case definition was generated from the tree-based machine learning model output combined with a feature importance algorithm. Two independent samples were used: one for training / testing the machine learning algorithm and the other for case definition validation.ResultsWe randomly selected 2,776 women aged 45–60 for this analysis and created a case definition, consisting of two occurrences within 24 months of International Classification of Diseases, Ninth Revision, Clinical Modification code 627 (or any sub-codes) OR one occurrence of Anatomical Therapeutic Chemical classification code G03CA (or any sub-codes) within the patient chart, that was highly effective at detecting problematic menopause cases. This definition produced a sensitivity of 81.5% (95% CI: 76.3-85.9%), specificity of 93.5% (91.9-94.8%), positive predictive value of 73.8% (68.3-78.6%), and negative predictive value of 95.7% (94.4-96.8%).ConclusionOur case definition for problematic menopause demonstrated high validity metrics and so is expected to be useful for epidemiological study and surveillance. This case definition will enable future studies exploring the management of menopause in primary care settings.