Incorporating Domain Knowledge Research Articles

On the Effectiveness of Large Language Models in Domain-Specific Code Generation

Large language models (LLMs) such as ChatGPT have shown remarkable capabilities in code generation. Despite significant achievements, they rely on enormous training data to acquire a broad spectrum of open-domain knowledge. Besides, their evaluation revolves around open-domain benchmarks like HumanEval, which primarily consist of programming contests. Therefore, it is hard to fully characterize the intricacies and challenges associated with particular domains (e.g., web, game, and math). In this paper, we conduct an in-depth study of the LLMs in domain-specific code generation. Our results demonstrate that LLMs exhibit sub-optimal performance in generating domain-specific code, due to their limited proficiency in utilizing domain-specific libraries. We further observe that incorporating API knowledge as prompts can empower LLMs to generate more professional code. Based on these findings, we further investigate how to effectively incorporate API knowledge into the code generation process. We experiment with three strategies for incorporating domain knowledge, namely, external knowledge inquirer, chain-of-thought prompting, and chain-of-thought fine-tuning. We refer to these strategies as a new code generation approach called DomCoder . Experimental results show that all strategies of DomCoder improve the effectiveness of domain-specific code generation under certain settings.

Bayesian material flow analysis for systems with multiple levels of disaggregation and high dimensional data

AbstractMaterial flow analysis (MFA) is used to quantify and understand the life cycles of materials from production to end of use, which enables environmental, social, and economic impacts and interventions. MFA is challenging as available data are often limited and uncertain, leading to an under‐determined system with an infinite number of possible stocks and flows values. Bayesian statistics is an effective way to address these challenges by principally incorporating domain knowledge, quantifying uncertainty in the data, and providing probabilities associated with model solutions. This paper presents a novel MFA methodology under the Bayesian framework. By relaxing the mass balance constraints, we improve the computational scalability and reliability of the posterior samples compared to existing Bayesian MFA methods. We propose a mass‐based, child and parent process framework to model systems with disaggregated processes and flows. We show posterior predictive checks can be used to identify inconsistencies in the data and aid noise and hyperparameter selection. The proposed approach is demonstrated in case studies, including a global aluminum cycle with significant disaggregation, under weakly informative priors and significant data gaps to investigate the feasibility of Bayesian MFA. We illustrate that just a weakly informative prior can greatly improve the performance of Bayesian methods, for both estimation accuracy and uncertainty quantification.

Explainable machine learning for assessing upper respiratory tract of racehorses from endoscopy videos

Laryngeal hemiplegia (LH) is a major upper respiratory tract (URT) complication in racehorses. Endoscopy imaging of horse throat is a gold standard for URT assessment. However, current manual assessment faces several challenges, stemming from the poor quality of endoscopy videos and subjectivity of manual grading. To overcome such limitations, we propose an explainable machine learning (ML)-based solution for efficient URT assessment. Specifically, a cascaded YOLOv8 architecture is utilized to segment the key semantic regions and landmarks per frame. Several spatiotemporal features are then extracted from key landmarks points and fed to a decision tree (DT) model to classify LH as Grade 1,2,3 or 4 denoting absence of LH, mild, moderate, and severe LH, respectively. The proposed method, validated through 5-fold cross-validation on 107 videos, showed promising performance in classifying different LH grades with 100%, 91.18%, 94.74% and 100% sensitivity values for Grade 1 to 4, respectively. Further validation on an external dataset of 72 cases confirmed its generalization capability with 90%, 80.95%, 100%, and 100% sensitivity values for Grade 1 to 4, respectively. We introduced several explainability related assessment functions, including: (i) visualization of YOLOv8 output to detect landmark estimation errors which can affect the final classification, (ii) time-series visualization to assess video quality, and (iii) backtracking of the DT output to identify borderline cases. We incorporated domain knowledge (e.g., veterinarian diagnostic procedures) into the proposed ML framework. This provides an assistive tool with clinical-relevance and explainability that can ease and speed up the URT assessment by veterinarians.

Behavior-aware assessment of demand response participation in power systems using causal artificial intelligence

Demand response (DR) is pivotal in enhancing the operational efficiency and reliability of power systems. Understanding prosumers’ behavior and their participation in DR programs is essential for utilities to design effective DR programs. Typically, prosumers are not concurrently exposed to multiple electricity price plans or engaged in distinct DR programs, which make the available data limited and introduces complexity in assessing their DR participation. Causal artificial intelligence offers a solution by enabling causal intervention and counterfactual analysis, facilitating the estimation of prosumers’ responses to demand-side programs without the necessity of experimental and control groups. This paper introduces a framework that integrates causal learning approaches for causal intervention and counterfactual analyses of prosumers’ DR participation and consumption behavior. The proposed framework incorporates domain knowledge of the power system to enhance model accuracy and performance. Simulation results demonstrate the effectiveness of the proposed framework with causal learning models in estimating prosumers’ DR participation and consumption behavior under different prices or DR programs. This capability allows utilities to gain a profound understanding of their prosumers, offering valuable insights for designing DR programs. Ultimately, this knowledge contributes to the reliable and efficient operation of power systems.

Assessment of living quality in Guangdong: A hybrid knowledge-based and data-driven approach

Air, water, and radiation in urban living environments are crucial factors affecting human health, societal well-being, and sustainable development. This study developed a novel hybrid knowledge-based and data-driven approach for analyzing air, water, and radiation monitoring data to assess the quality of urban living natural environments (ULNEQ) in Guangdong Province. Fuzzy set-pair analysis was employed for data preprocessing, effectively incorporating domain knowledge into the raw data. Then, a hierarchical clustering algorithm was utilized to evaluate the ULNEQ level. The proposed approach enhanced data interpretability and optimized the clustering process, yielding more robust and reliable analytical results. The results revealed a diverse environmental landscape across the province, with Heyuan and Meizhou consistently maintaining high standards, whereas Dongguan and Jieyang exhibited notably poor conditions, ranking at Level IV or V for over 72% of the observed months. Key factors such as monthly average concentration ozone (0.56), the city water quality index (0.49), and the proportion of days with standard air quality (PDSAQ) (−0.39) significantly influenced the ULNEQ, with PDSAQ showing a negative correlation. Notably, a province-wide observation in June 2022 showed all cities maintaining ULEQ gradings of Level II or better. This innovative approach can be adapted to enhance the management and control of ULNEQ.

Intrinsically interpretable machine learning-based building energy load prediction method with high accuracy and strong interpretability

Black-box models have demonstrated remarkable accuracy in forecasting building energy loads. However, they usually lack interpretability and do not incorporate domain knowledge, making it difficult for users to trust their predictions in practical applications. One important and interesting question remains unanswered: is it possible to use intrinsically interpretable models to achieve accuracy comparable to that of black-box models? With an aim of answering this question, this study proposes an intrinsically interpretable machine learning-based method to forecast building energy loads. It creatively combines two intrinsically interpretable machine learning algorithms: clustering decision trees and adaptive multiple linear regression. Clustering decision trees aim to automatically identify various building operation conditions, allowing for the training of multiple models tailored to each condition. It can reduce the complexity of model training data, leading to higher accuracy. Adaptive multiple linear regression is an improved regression algorithm tailored to building energy load prediction. It can adaptively modify regression coefficients according to building operations, enhancing the non-linear fitting capability of multiple linear regression. The proposed method is evaluated utilizing the operational data from an office building. The results indicate that the proposed method exhibits comparable accuracy to both random forests and extreme gradient boosting. Furthermore, it shows significantly superior accuracy, with an average improvement of 10.2 %, compared with some popular black-box algorithms such as artificial neural networks, support vector regression, and classification and regression trees. As for model interpretability, the proposed method reveals that historical cooling loads are the most crucial for predicting building cooling loads under most conditions. Additionally, outdoor air temperature has a significant contribution to building cooling load prediction during the daytime on weekdays in summer and transition seasons. In the future, it will be valuable to explore integrating the laws of physics into the proposed method to further enhance its interpretability.

Efficient Clothing Fashion Prediction using Machine Learning

Fashion trend forecasting is a crucial task for both academia and industry. Although some efforts have been devoted to tackling this challenging task, they only studied limited fashion elements with highly seasonal or simple patterns, which could hardly reveal the real fashion trends. Towards insightful fashion trend forecasting, this work focuses on investigating fine-grained fashion element trends for specific user groups. We first contribute a large-scale fashion trend dataset (FIT) collected from social media with extracted time series fashion element records and user information. Furthermore, to effectively model the time series data of fashion elements with rather complex patterns, we propose a Machine Learning which takes advantage of the capability in modeling time series data. Moreover, it leverages internal and external knowledge in fashion domain that affects the time-series patterns of fashion element trends. Such incorporation of domain knowledge further enhances the deep learning model in capturing the patterns of specific fashion elements and predicting the future trends. Extensive experiments demonstrate that the proposed ML model can effectively capture the complicated patterns of objective fashion elements, therefore making preferable fashion trend forecast.

Learning Robust Shape Regularization for Generalizable Medical Image Segmentation.

Generalizable medical image segmentation enables models to generalize to unseen target domains under domain shift issues. Recent progress demonstrates that the shape of the segmentation objective, with its high consistency and robustness across domains, can serve as a reliable regularization to aid the model for better cross-domain performance, where existing methods typically seek a shared framework to render segmentation maps and shape prior concurrently. However, due to the inherent texture and style preference of modern deep neural networks, the edge or silhouette of the extracted shape will inevitably be undermined by those domain-specific texture and style interferences of medical images under domain shifts. To address this limitation, we devise a novel framework with a separation between the shape regularization and the segmentation map. Specifically, we first customize a novel whitening transform-based probabilistic shape regularization extractor namely WT-PSE to suppress undesirable domain-specific texture and style interferences, leading to more robust and high-quality shape representations. Second, we deliver a Wasserstein distance-guided knowledge distillation scheme to help the WT-PSE to achieve more flexible shape extraction during the inference phase. Finally, by incorporating domain knowledge of medical images, we propose a novel instance-domain whitening transform method to facilitate a more stable training process with improved performance. Experiments demonstrate the performance of our proposed method on both multi-domain and single-domain generalization.

Contingency analysis with warm starter using probabilistic graphical model

Cyberthreats are an increasingly common risk to the power grid and can thwart secure grid operations. We propose to extend contingency analysis to include cyberthreat evaluations. However, unlike the traditional N-1 or N-2 contingencies, cyberthreats (e.g., MadIoT) require simulating hard-to-solve N-k (with k ≫ 2) contingencies in a practical amount of time. Purely physics-based power flow solvers, while being accurate, are slow and may not solve N-k contingencies in a timely manner, whereas the emerging data-driven alternatives are fast but not sufficiently generalizable, interpretable, and scalable. To address these challenges, we propose a novel conditional Gaussian Random Field-based data-driven method that performs fast and accurate evaluation of cyberthreats. It achieves speedup of contingency analysis by warm-starting simulations, i.e., improving starting points, for the physical solvers. To improve the physical interpretability and generalizability, the proposed method incorporates domain knowledge by considering the graphical nature of the grid topology. To improve scalability, the method applies physics-informed regularization that reduces model complexity. Experiments validate that simulating MadIoT-induced attacks with our warm starter becomes approximately 5x faster on a realistic 2000-bus system.

Comparative Analysis of Machine Learning Models for Chronic Disease Indicator Classification Using U.S. Chronic Disease Indicators Dataset

The prevalence of chronic diseases poses significant challenges to public health systems worldwide. This study evaluates the performance of four machine learning models—Gradient Boosting Classifier, Support Vector Machine (SVM), Logistic Regression, and Random Forest—in classifying chronic disease indicators using the U.S. Chronic Disease Indicators (CDI) dataset. The models were assessed based on accuracy, precision, recall, F1 score, classification report, and confusion matrix to determine their effectiveness. The Gradient Boosting Classifier outperformed other models with an accuracy of 64.36%, precision of 63.72%, recall of 64.36%, and F1 score of 63.88%. While SVM and Random Forest demonstrated moderate performance, Logistic Regression served as a baseline for comparison. The study highlights the Gradient Boosting Classifier's superiority in handling the complexities of the CDI dataset, suggesting its potential for improving chronic disease prediction and management. Future research should focus on refining these models, addressing class imbalances, and incorporating domain knowledge to enhance interpretability and applicability in real-world scenarios.

A generic causality-informed neural network (CINN) methodology for quantitative risk analytics and decision support.

In this paper, we develop a generic framework for systemically encoding causal knowledge manifested in the form of hierarchical causality structure and qualitative (or quantitative) causal relationships into neural networks to facilitate sound risk analytics and decision support via causally-aware intervention reasoning. The proposed methodology for establishing causality-informed neural network (CINN) follows a four-step procedure. In the first step, we explicate how causal knowledge in the form of directed acyclic graph (DAG) can be discovered from observation data or elicited from domain experts. Next, we categorize nodes in the constructed DAG representing causal relationships among observed variables into several groups (e.g., root nodes, intermediate nodes, and leaf nodes), and align the architecture of CINN with causal relationships specified in the DAG while preserving the orientation of each existing causal relationship. In addition to a dedicated architecture design, CINN also gets embodied in the design of loss function, where both intermediate and leaf nodes are treated as target outputs to be predicted by CINN. In the third step, we propose to incorporate domain knowledge on stable causal relationships into CINN, and the injected constraints on causal relationships act as guardrails to prevent unexpected behaviors of CINN. Finally, the trained CINN is exploited to perform intervention reasoning with emphasis on estimating the effect that policies and actions can have on the system behavior, thus facilitating risk-informed decision making through comprehensive "what-if" analysis. Two case studies are used to demonstrate the substantial benefits enabled by CINN in risk analytics and decisionsupport.

A Novel Hybrid Method for Predicting Specific Crop Yield

Predicting crop yields is crucial for economic planning, food security, and agricultural management. Conventional approaches frequently need accurate results regarding the complex interaction between environmental conditions and crop-specific traits. By combining the best features of machine learning algorithms with expert-level domain knowledge, we provide a new hybrid approach to predicting crop yields that is both state- and crop-specific. Our approach integrates historical yield data, satellite imagery, meteorological data, soil properties, and crop-specific information to build robust predictive models. Specifically, we employ ensemble learning techniques, including random forest and gradient boosting, to capture complex nonlinear relationships and improve prediction accuracy. Additionally, we incorporate domain knowledge through feature engineering and selection to enhance model interpretability and generalization. We validate our method using real-world datasets from diverse geographical regions and crops, demonstrating its superior performance compared to traditional approaches. Our proposed hybrid method offers a promising solution for accurate and reliable crop yield prediction, facilitating informed agricultural decision-making.

МЕТОД ІНТЕГРУВАННЯ ДОМЕННИХ ЗНАНЬ У БАГАТОСТРАТЕГІЧНУ КЛАСИФІКАЦІЮ МЕДИЧНИХ ЗОБРАЖЕНЬ

The integration of domain-specific knowledge into medical image classification systems presents a promising approach to enhance the accuracy and interpretability of diagnostic models, particularly in the field of medical imaging. This study aimed to evaluate the efficacy of incorporating domain knowledge into multi-strategic classification algorithms for medical images, with a focus on magnetic resonance imaging (MRI) of the brain. This was a comparative, explorative study utilizing a newly developed multi-strategic classification framework that integrates medical domain knowledge into the classification process. The dataset comprised MRI scans of the brain, collected from three different medical institutions, totaling 320 images, which were annotated by expert radiologists. The framework employed basic classifiers, where each classifier was trained on examples incorrectly classified by another, fostering a collaborative learning environment. Performance was compared against traditional single-strategy classification methods. The integrated multi-strategic classification model demonstrated superior performance, achieving higher sensitivity, specificity, and accuracy with 98 %, 97 %, and 96 %, respectively. Furthermore, the model significantly improved the interpretability of diagnostic predictions, aligning closely with expert radiologists' assessments. The incorporation of domain-specific knowledge into the classification of medical images significantly enhances both the accuracy and the interpretability of the results. The multi-strategic approach allows for a more robust and nuanced understanding of complex medical images, such as MRI scans of the brain, potentially leading to improved diagnostic processes and outcomes in clinical settings. Future research should focus on expanding the types of medical images analyzed and further refining the integration techniques to encompass a broader spectrum of domain knowledge.

NREM3 Sleep Stage Estimation Based on Accelerometer by Body Movement Count and Biological Rhythms

This paper proposes the method by physiological knowledge to improve the estimation performance of the NREM3 sleep based on the waist-attached accelerometer. Specifically, this paper proposes the hybrid method that combines the method based on body movement counts and the method based on biological rhythms of sleep. Through the human subject experiment, the following implications were revealed: (1) the proposed method can outperform famous machine learning models (Random Forest and LSTM) trained with automatically generated features that do not sufficiently incorporate domain knowledge; (2) when the input features are based on domain knowledge, the estimator explicitly designed by humans can outperform the machine learning method; and (3) combining the body movement counting method and the biological rhythm-based method can suppress the error of the body movement counting method and reduce false positives.

ADHD diagnosis guided by functional brain networks combined with domain knowledge

Utilizing functional magnetic resonance imaging (fMRI) to model functional brain networks (FBNs) is increasingly prominent in attention-deficit/hyperactivity disorder (ADHD) research, revealing neural impact and mechanisms through the exploration of activated brain regions. However, current FBNs-based methods face two major challenges. The primary challenge stems from the limitations of existing modeling methods in accurately capturing both regional correlations and long-distance dependencies (LDDs) within the dynamic brain, thereby affecting the diagnostic accuracy of FBNs as biomarkers. Additionally, limited sample size and class imbalance also pose a challenge to the learning performance of the model. To address the issues, we propose an automated diagnostic framework, which integrates modeling, multimodal fusion, and classification into a unified process. It aims to extract representative FBNs and efficiently incorporate domain knowledge to guide ADHD classification. Our work mainly includes three-fold: 1) A multi-head attention-based region-enhancement module (MAREM) is designed to simultaneously capture regional correlations and LDDs across the entire sequence of brain activity, which facilitates the construction of representative FBNs. 2) The multimodal supplementary learning module (MSLM) is proposed to integrate domain knowledge from phenotype data with FBNs from neuroimaging data, achieving information complementarity and alleviating the problems of insufficient medical data and unbalanced sample categories. 3) An ADHD automatic diagnosis framework guided by FBNs and domain knowledge (ADF-FAD) is proposed to help doctors make more accurate decisions, which is applied to the ADHD-200 dataset to confirm its effectiveness. The results indicate that the FBNs extracted by MAREM perform well in modeling and classification. After with MSLM, the model achieves accuracy of 92.4%, 74.4%, and 80% at NYU, PU, and KKI, respectively, demonstrating its ability to effectively capture crucial information related to ADHD diagnosis. Codes are available at https://github.com/zhuimengxuebao/ADF-FAD.

Analysis of Missingness Scenarios for Observational Health Data.

Despite the extensive literature on missing data theory and cautionary articles emphasizing the importance of realistic analysis for healthcare data, a critical gap persists in incorporating domain knowledge into the missing data methods. In this paper, we argue that the remedy is to identify the key scenarios that lead to data missingness and investigate their theoretical implications. Based on this proposal, we first introduce an analysis framework where we investigate how different observation agents, such as physicians, influence the data availability and then scrutinize each scenario with respect to the steps in the missing data analysis. We apply this framework to the case study of observational data in healthcare facilities. We identify ten fundamental missingness scenarios and show how they influence the identification step for missing data graphical models, inverse probability weighting estimation, and exponential tilting sensitivity analysis. To emphasize how domain-informed analysis can improve method reliability, we conduct simulation studies under the influence of various missingness scenarios. We compare the results of three common methods in medical data analysis: complete-case analysis, Missforest imputation, and inverse probability weighting estimation. The experiments are conducted for two objectives: variable mean estimation and classification accuracy. We advocate for our analysis approach as a reference for the observational health data analysis. Beyond that, we also posit that the proposed analysis framework is applicable to other medical domains.

A domain knowledge-informed design space exploration methodology for mechanical layout design

Layout designs have a large number of design variables and various physical constraints. Conventional design exploration approaches are time-consuming and may require human intervention. A unique feature about physical layout designs is the availability of domain knowledge, which can be utilised to speed up the design process. In this paper, we propose a generative deep learning-based design space exploration (DSE) methodology that is capable of learning design constraints in the layout design without explicit supervision. Moreover, it can incorporate domain knowledge in the generated layouts, thereby speeding up the design process. This is realised by constructing a layout generation variational autoencoder (LGVAE) model, which uses a latent space as an interface to generate the layouts. By training the LGVAE model, significantly lower-dimensional representations can be learned compared to the original dimensionality of the design space. Therefore, the number of design variables is greatly reduced. We showcase the performance of the proposed DSE approach by solving the heat source layout design problem encountered in thermal management of chips. Experiments demonstrate that the LGVAE model is capable of generating compressed latent representations that capture the characteristics of the input samples, which makes the DSE cost-effective.

Physics-Informed Transfer Learning to Enhance Sleep Staging.

At-home sleep staging using wearable medical sensors poses a viable alternative to in-hospital polysomnography due to its lower cost and lower disruption to the daily lives of patients, especially in the case of long-term monitoring. Machine learning with wearables however is difficult due to the paucity of data from wearable sensors, making automation a challenge. Transfer learning from hospital polysomnograms can boost performance, but is still hindered by differences between wearable and in-hospital EEG resulting in part from differing electrode placement. We improve transfer learning performance by using electrophysiological models of a human head to generate synthetic EEG resembling EEG from a wearable sensor. The data generation method utilizes Low-Resolution Electromagnetic Tomography Analysis (LORETA). Real EEG from standard in- hospital recordings is first mapped to point currents within the brain using LORETA, after which the point currents are used to estimate EEG that would have been recorded using a wearable sensor at any given point on the head. Augmenting source datasets with synthetic data statistically significantly boosted accuracy on a wearable sleep staging task from 80.8% to 81.3% on average, depending on the transfer learning parameters and data sources. Machine learning performance can be improved using data synthesized using physical models. Our approach represents a new form of transfer learning and demonstrates that incorporating domain knowledge of electrophysiological modeling can improve machine learning results for sleep staging tasks. We expect this approach to be particularly useful for EEG data which is hard to collect, or which is obtained using unusual electrode configurations.

Application of deep learning model incorporating domain knowledge in international migration forecasting

PurposeForecasting population movement trends is crucial for implementing effective policies to regulate labor force growth and understand demographic changes. Combining migration theory expertise and neural network technology can bring a fresh perspective to international migration forecasting research.Design/methodology/approachThis study proposes a conditional generative adversarial neural network model incorporating the migration knowledge – conditional generative adversarial network (MK-CGAN). By using the migration knowledge to design the parameters, MK-CGAN can effectively address the limited data problem, thereby enhancing the accuracy of migration forecasts.FindingsThe model was tested by forecasting migration flows between different countries and had good generalizability and validity. The results are robust as the proposed solutions can achieve lesser mean absolute error, mean squared error, root mean square error, mean absolute percentage error and R2 values, reaching 0.9855 compared to long short-term memory (LSTM), gated recurrent unit, generative adversarial network (GAN) and the traditional gravity model.Originality/valueThis study is significant because it demonstrates a highly effective technique for predicting international migration using conditional GANs. By incorporating migration knowledge into our models, we can achieve prediction accuracy, gaining valuable insights into the differences between various model characteristics. We used SHapley Additive exPlanations to enhance our understanding of these differences and provide clear and concise explanations for our model predictions. The results demonstrated the theoretical significance and practical value of the MK-CGAN model in predicting international migration.

Domain Knowledge-Driven Relation Extraction Methods

Relation classification is one of the important tasks in natural language processing, which aims to determine the relationship between entities given a sentence and their respective positions. Most of the existing methods for relation classification are based on neural networks using pre-trained models such as BERT. In recent years, models based on pre-trained models like BERT have achieved excellent results in relation classification tasks in general domains. However, these methods often struggle when applied to specialized domains due to the limitations of the corpora used for BERT pre-training. Most pre-trained models are trained on text corpora from general sources like Wikipedia, which cover a wide range of domains. As a result, the content of these corpora in specific domains is limited and lacks the necessary expertise, leading to subpar performance of relation classification models in specific domains. While providing a large number of domain-specific corpora to pre-trained models could potentially address this issue, it comes with limitations such as increased computational requirements and insufficient training of specialized vocabulary. This paper proposes a method inspired by the K-BERT pre-training model to incorporate triplet knowledge from domain knowledge graphs into sentence sequences. The triplets are transformed into sentence trees and then fed into the BERT pre-trained model using absolute and relative indices. Our model has achieved an accuracy of 93.06%, which is markedly higher than that of any other baseline approach. This approach allows the incorporation of domain knowledge without significantly increasing the computational complexity. Additionally, the paper introduces a partial input method that enables the computer to understand input sentences from multiple dimensions and hierarchical levels. Experimental results on a medical domain dataset for relation classification, which includes type labels, demonstrate significant advantages over other relation classification models in terms of Accuracy.