The development of Generative AI Large Language Models (LLMs) raised the alarm regarding identifying content produced through generative AI or humans. In one case, issues arise when students heavily rely on such tools in a manner that can affect the development of their writing or coding skills. Other issues of plagiarism also apply. This study aims to support efforts to detect and identify textual content generated using LLM tools. We hypothesize that LLMs-generated text is detectable by machine learning (ML), and investigate ML models that can recognize and differentiate texts generated by multiple LLMs tools. We leverage several ML and Deep Learning (DL) algorithms such as Random Forest (RF), and Recurrent Neural Networks (RNN), and utilized Explainable Artificial Intelligence (XAI) to understand the important features in attribution. Our method is divided into 1) binary classification to differentiate between human-written and AI-text, and 2) multi classification, to differentiate between human-written text and the text generated by the five different LLM tools (ChatGPT, LLaMA, Google Bard, Claude, and Perplexity). Results show high accuracy in the multi and binary classification. Our model outperformed GPTZero with 98.5\% accuracy to 78.3\%. Notably, GPTZero was unable to recognize about 4.2\% of the observations, but our model was able to recognize the complete test dataset. XAI results showed that understanding feature importance across different classes enables detailed author/source profiles. Further, aiding in attribution and supporting plagiarism detection by highlighting unique stylistic and structural elements ensuring robust content originality verification.
We propose Group Shapley, a metric that extends the classical individual-level Shapley value framework to evaluate the importance of feature groups, addressing the structured nature of predictors commonly found in business and economic data. More importantly, we develop a significance testing procedure based on a three-cumulant chi-square approximation and establish the asymptotic properties of the test statistics for Group Shapley values. Our approach can effectively handle challenging scenarios, including sparse or skewed distributions and small sample sizes, outperforming alternative tests such as the Wald test. Simulations confirm that the proposed test maintains robust empirical size and demonstrates enhanced power under diverse conditions. To illustrate the method's practical relevance in advancing Explainable AI, we apply our framework to bond recovery rate predictions using a global dataset (1996-2023) comprising 2,094 observations and 98 features, grouped into 16 subgroups and five broader categories: bond characteristics, firm fundamentals, industry-specific factors, market-related variables, and macroeconomic indicators. Our results identify the market-related variables group as the most influential. Furthermore, Lorenz curves and Gini indices reveal that Group Shapley assigns feature importance more equitably compared to individual Shapley values.
Humour styles can have either a negative or a positive impact on well-being. Given the importance of these styles to mental health, significant research has been conducted on their automatic identification. However, the automated machine learning models used for this purpose are black boxes, making their prediction decisions opaque. Clarity and transparency are vital in the field of mental health. This paper presents an explainable AI (XAI) framework for understanding humour style classification, building upon previous work in computational humour analysis. Using the best-performing single model (ALI+XGBoost) from prior research, we apply comprehensive XAI techniques to analyse how linguistic, emotional, and semantic features contribute to humour style classification decisions. Our analysis reveals distinct patterns in how different humour styles are characterised and misclassified, with particular emphasis on the challenges in distinguishing affiliative humour from other styles. Through detailed examination of feature importance, error patterns, and misclassification cases, we identify key factors influencing model decisions, including emotional ambiguity, context misinterpretation, and target identification. The framework demonstrates significant utility in understanding model behaviour, achieving interpretable insights into the complex interplay of features that define different humour styles. Our findings contribute to both the theoretical understanding of computational humour analysis and practical applications in mental health, content moderation, and digital humanities research.
Counterfactual examples are widely used in natural language processing (NLP) as valuable data to improve models, and in explainable artificial intelligence (XAI) to understand model behavior. The automated generation of counterfactual examples remains a challenging task even for large language models (LLMs), despite their impressive performance on many tasks. In this paper, we first introduce ZeroCF, a faithful approach for leveraging important words derived from feature attribution methods to generate counterfactual examples in a zero-shot setting. Second, we present a new framework, FitCF, which further verifies aforementioned counterfactuals by label flip verification and then inserts them as demonstrations for few-shot prompting, outperforming two state-of-the-art baselines. Through ablation studies, we identify the importance of each of FitCF's core components in improving the quality of counterfactuals, as assessed through flip rate, perplexity, and similarity measures. Furthermore, we show the effectiveness of LIME and Integrated Gradients as backbone attribution methods for FitCF and find that the number of demonstrations has the largest effect on performance. Finally, we reveal a strong correlation between the faithfulness of feature attribution scores and the quality of generated counterfactuals.
High-frequency trading (HFT) has transformed modern financial markets, making reliable short-term price forecasting models essential. In this study, we present a novel approach to mid-price forecasting using Level 1 limit order book (LOB) data from NASDAQ, focusing on 100 U.S. stocks from the S&P 500 index during the period from September to November 2022. Expanding on our previous work with Radial Basis Function Neural Networks (RBFNN), which leveraged automated feature importance techniques based on mean decrease impurity (MDI) and gradient descent (GD), we introduce the Adaptive Learning Policy Engine (ALPE) - a reinforcement learning (RL)-based agent designed for batch-free, immediate mid-price forecasting. ALPE incorporates adaptive epsilon decay to dynamically balance exploration and exploitation, outperforming a diverse range of highly effective machine learning (ML) and deep learning (DL) models in forecasting performance.
Underwater Image Restoration (UIR) remains a challenging task in computer vision due to the complex degradation of images in underwater environments. While recent approaches have leveraged various deep learning techniques, including Transformers and complex, parameter-heavy models to achieve significant improvements in restoration effects, we demonstrate that pure CNN architectures with lightweight parameters can achieve comparable results. In this paper, we introduce UIR-PolyKernel, a novel method for underwater image restoration that leverages Polymorphic Large Kernel CNNs. Our approach uniquely combines large kernel convolutions of diverse sizes and shapes to effectively capture long-range dependencies within underwater imagery. Additionally, we introduce a Hybrid Domain Attention module that integrates frequency and spatial domain attention mechanisms to enhance feature importance. By leveraging the frequency domain, we can capture hidden features that may not be perceptible to humans but are crucial for identifying patterns in both underwater and on-air images. This approach enhances the generalization and robustness of our UIR model. Extensive experiments on benchmark datasets demonstrate that UIR-PolyKernel achieves state-of-the-art performance in underwater image restoration tasks, both quantitatively and qualitatively. Our results show that well-designed pure CNN architectures can effectively compete with more complex models, offering a balance between performance and computational efficiency. This work provides new insights into the potential of CNN-based approaches for challenging image restoration tasks in underwater environments. The code is available at \href{https://github.com/CXH-Research/UIR-PolyKernel}{https://github.com/CXH-Research/UIR-PolyKernel}.
This study presents an autonomous experimental machine learning protocol for high-frequency trading (HFT) stock price forecasting that involves a dual competitive feature importance mechanism and clustering via shallow neural network topology for fast training. By incorporating the k-means algorithm into the radial basis function neural network (RBFNN), the proposed method addresses the challenges of manual clustering and the reliance on potentially uninformative features. More specifically, our approach involves a dual competitive mechanism for feature importance, combining the mean-decrease impurity (MDI) method and a gradient descent (GD) based feature importance mechanism. This approach, tested on HFT Level 1 order book data for 20 S&P 500 stocks, enhances the forecasting ability of the RBFNN regressor. Our findings suggest that an autonomous approach to feature selection and clustering is crucial, as each stock requires a different input feature space. Overall, by automating the feature selection and clustering processes, we remove the need for manual topological grid search and provide a more efficient way to predict LOB's mid-price.
Because of the theoretical challenges posed by the Efficient Market Hypothesis to technical analysis, the effectiveness of technical indicators in high-frequency trading remains inadequately explored, particularly at the minute-level frequency, where effects of the microstructure of the market dominate. This study evaluates the integration of traditional technical indicators with random forest regression models using minute-level SPY data, analyzing 13 distinct model configurations. Our empirical results reveal a stark contrast between in-sample and out-of-sample performance, with $R^2$ values deteriorating from 0.749--0.812 during training to negative values in testing. A feature importance analysis demonstrates that primary price-based features dominate the predictions made by the model, accounting for over 60% of the importance, while established technical indicators, such as RSI and Bollinger Bands, account for only 14%--15%. Although the indicator-enhanced models achieved superior risk-adjusted metrics, with Rachev ratios between 0.919 and 0.961, they consistently underperformed a simple buy-and-hold strategy, generating returns ranging from -2.4% to -3.9%. These findings challenge conventional assumptions about the usefulness of technical indicators in algorithmic trading, suggesting that in high-frequency contexts, they may be more relevant to risk management rather than to predicting returns. For practitioners and researchers, our findings indicate that successful high-frequency trading strategies should focus on adaptive feature selection and regime-specific modeling rather than relying on traditional technical indicators, as well as indicating the critical importance of robust out-of-sample testing in the development of a model.
Machine Learning is transforming medical research by improving diagnostic accuracy and personalizing treatments. General ML models trained on large datasets identify broad patterns across populations, but their effectiveness is often limited by the diversity of human biology. This has led to interest in subject-specific models that use individual data for more precise predictions. However, these models are costly and challenging to develop. To address this, we propose a novel validation approach that uses a general ML model to ensure reproducible performance and robust feature importance analysis at both group and subject-specific levels. We tested a single Random Forest (RF) model on nine datasets varying in domain, sample size, and demographics. Different validation techniques were applied to evaluate accuracy and feature importance consistency. To introduce variability, we performed up to 400 trials per subject, randomly seeding the ML algorithm for each trial. This generated 400 feature sets per subject, from which we identified top subject-specific features. A group-specific feature importance set was then derived from all subject-specific results. We compared our approach to conventional validation methods in terms of performance and feature importance consistency. Our repeated trials approach, with random seed variation, consistently identified key features at the subject level and improved group-level feature importance analysis using a single general model. Subject-specific models address biological variability but are resource-intensive. Our novel validation technique provides consistent feature importance and improved accuracy within a general ML model, offering a practical and explainable alternative for clinical research.
Tabular data is a common format for storing information in rows and columns to represent data entries and their features. Although deep neural networks have become the main approach for modeling a wide range of domains including computer vision and NLP, many of them are not well-suited for tabular data. Recently, a few deep learning models have been proposed for deep tabular learning, featuring an internal feature selection mechanism with end-to-end gradient-based optimization. However, their feature selection mechanisms are unidimensional, and hence fail to account for the contextual dependence of feature importance, potentially overlooking crucial interactions that govern complex tasks. In addition, they overlook the bias of high-impact features and the risk associated with the limitations of attention generalization. To address this limitation, this study proposes a novel iterative feature exclusion module that enhances the feature importance ranking in tabular data. The proposed module iteratively excludes each feature from the input data and computes the attention scores, which represent the impact of the features on the prediction. By aggregating the attention scores from each iteration, the proposed module generates a refined representation of feature importance that captures both global and local interactions between features. The effectiveness of the proposed module is evaluated on four public datasets. The results demonstrate that the proposed module consistently outperforms state-of-the-art methods and baseline models in feature ranking and classification tasks. The code is publicly available at https://github.com/abaraka2020/Iterative-Feature-Exclusion-Ranking-Module and https://github.com/mohalim/IFENet
Explaining machine learning (ML) models using eXplainable AI (XAI) techniques has become essential to make them more transparent and trustworthy. This is especially important in high-stakes domains like healthcare, where understanding model decisions is critical to ensure ethical, sound, and trustworthy outcome predictions. However, users are often confused about which explanability method to choose for their specific use case. We present a comparative analysis of widely used explainability methods, Shapley Additive Explanations (SHAP) and Gradient-weighted Class Activation Mapping (GradCAM), within the domain of human activity recognition (HAR) utilizing graph convolutional networks (GCNs). By evaluating these methods on skeleton-based data from two real-world datasets, including a healthcare-critical cerebral palsy (CP) case, this study provides vital insights into both approaches' strengths, limitations, and differences, offering a roadmap for selecting the most appropriate explanation method based on specific models and applications. We quantitatively and quantitatively compare these methods, focusing on feature importance ranking, interpretability, and model sensitivity through perturbation experiments. While SHAP provides detailed input feature attribution, GradCAM delivers faster, spatially oriented explanations, making both methods complementary depending on the application's requirements. Given the importance of XAI in enhancing trust and transparency in ML models, particularly in sensitive environments like healthcare, our research demonstrates how SHAP and GradCAM could complement each other to provide more interpretable and actionable model explanations.
Tree-based learning methods such as Random Forest and XGBoost are still the gold-standard prediction methods for tabular data. Feature importance measures are usually considered for feature selection as well as to assess the effect of features on the outcome variables in the model. This also applies to survey data, which are frequently encountered in the social sciences and official statistics. These types of datasets often present the challenge of missing values. The typical solution is to impute the missing data before applying the learning method. However, given the large number of possible imputation methods available, the question arises as to which should be chosen to achieve the 'best' reflection of feature importance and feature selection in subsequent analyses. In the present paper, we investigate this question in a survey-based simulation study for eight state-of-the art imputation methods and three learners. The imputation methods comprise listwise deletion, three MICE options, four \texttt{missRanger} options as well as the recently proposed mixGBoost imputation approach. As learners, we consider the two most common tree-based methods, Random Forest and XGBoost, and an interpretable linear model with regularization.
Time perception research has advanced significantly over the years. However, some areas remain largely unexplored. This study addresses two such under-explored areas in timing research: (1) A quantitative analysis of time perception at an individual level, and (2) Time perception in an ecological setting. In this context, we trained a machine learning model to predict the direction of change in an individual's time production. The model's training data was collected using an ecologically valid setup. We moved closer to an ecological setting by conducting an online experiment with 995 participants performing a time production task that used naturalistic videos (no audio) as stimuli. The model achieved an accuracy of 61%. This was 10 percentage points higher than the baseline models derived from cognitive theories of timing. The model performed equally well on new data from a second experiment, providing evidence of its generalization capabilities. The model's output analysis revealed that it also contained information about the magnitude of change in time production. The predictions were further analysed at both population and individual level. It was found that a participant's previous timing performance played a significant role in determining the direction of change in time production. By integrating attentional-gate theories from timing research with feature importance techniques from machine learning, we explained model predictions using cognitive theories of timing. The model and findings from this study have potential applications in systems involving human-computer interactions where understanding and predicting changes in user's time perception can enable better user experience and task performance.
This project aims to predict short-term and long-term upward trends in the S&P 500 index using machine learning models and feature engineering based on the "101 Formulaic Alphas" methodology. The study employed multiple models, including Logistic Regression, Decision Trees, Random Forests, Neural Networks, K-Nearest Neighbors (KNN), and XGBoost, to identify market trends from historical stock data collected from Yahoo! Finance. Data preprocessing involved handling missing values, standardization, and iterative feature selection to ensure relevance and variability. For short-term predictions, KNN emerged as the most effective model, delivering robust performance with high recall for upward trends, while for long-term forecasts, XGBoost demonstrated the highest accuracy and AUC scores after hyperparameter tuning and class imbalance adjustments using SMOTE. Feature importance analysis highlighted the dominance of momentum-based and volume-related indicators in driving predictions. However, models exhibited limitations such as overfitting and low recall for positive market movements, particularly in imbalanced datasets. The study concludes that KNN is ideal for short-term alerts, whereas XGBoost is better suited for long-term trend forecasting. Future enhancements could include advanced architectures like Long Short-Term Memory (LSTM) networks and further feature refinement to improve precision and generalizability. These findings contribute to developing reliable machine learning tools for market trend prediction and investment decision-making.
Leveraging the large body of work devoted in recent years to describe redundancy and synergy in multivariate interactions among random variables, we propose a novel approach to quantify cooperative effects in feature importance, one of the most used techniques for explainable artificial intelligence. In particular, we propose an adaptive version of a well-known metric of feature importance, named Leave One Covariate Out (LOCO), to disentangle high-order effects involving a given input feature in regression problems. LOCO is the reduction of the prediction error when the feature under consideration is added to the set of all the features used for regression. Instead of calculating the LOCO using all the features at hand, as in its standard version, our method searches for the multiplet of features that maximize LOCO and for the one that minimize it. This provides a decomposition of the LOCO as the sum of a two-body component and higher-order components (redundant and synergistic), also highlighting the features that contribute to building these high-order effects alongside the driving feature. We report the application to proton/pion discrimination from simulated detector measures by GEANT.
Feature selection is crucial in recommender systems for improving model efficiency and predictive performance. Traditional methods rely on agency models, such as decision trees or neural networks, to estimate feature importance. However, this approach is inherently limited, as the agency models may fail to learn effectively in all scenarios due to suboptimal training conditions (e.g., feature collinearity, high-dimensional sparsity, and data insufficiency). In this paper, we propose AltFS, an Agency-light Feature Selection method for deep recommender systems. AltFS integrates semantic reasoning from Large Language Models (LLMs) with task-specific learning from agency models. Initially, LLMs will generate a semantic ranking of feature importance, which is then refined by an agency model, combining world knowledge with task-specific insights. Extensive experiments on three public datasets from real-world recommender platforms demonstrate the effectiveness of AltFS. Our code is publicly available for reproducibility.
State-of-the-art Active Speaker Detection (ASD) approaches mainly use audio and facial features as input. However, the main hypothesis in this paper is that body dynamics is also highly correlated to "speaking" (and "listening") actions and should be particularly useful in wild conditions (e.g., surveillance settings), where face cannot be reliably accessed. We propose ASDnB, a model that singularly integrates face with body information by merging the inputs at different steps of feature extraction. Our approach splits 3D convolution into 2D and 1D to reduce computation cost without loss of performance, and is trained with adaptive weight feature importance for improved complement of face with body data. Our experiments show that ASDnB achieves state-of-the-art results in the benchmark dataset (AVA-ActiveSpeaker), in the challenging data of WASD, and in cross-domain settings using Columbia. This way, ASDnB can perform in multiple settings, which is positively regarded as a strong baseline for robust ASD models (code available at https://github.com/Tiago-Roxo/ASDnB).
Understanding how changes in explanatory features affect the unconditional distribution of the outcome is important in many applications. However, existing black-box predictive models are not readily suited for analyzing such questions. In this work, we develop an approximation method to compute the feature importance curves relevant to the unconditional distribution of outcomes, while leveraging the power of pre-trained black-box predictive models. The feature importance curves measure the changes across quantiles of outcome distribution given an external impact of change in the explanatory features. Through extensive numerical experiments and real data examples, we demonstrate that our approximation method produces sparse and faithful results, and is computationally efficient.
State-of-the-art Active Speaker Detection (ASD) approaches heavily rely on audio and facial features to perform, which is not a sustainable approach in wild scenarios. Although these methods achieve good results in the standard AVA-ActiveSpeaker set, a recent wilder ASD dataset (WASD) showed the limitations of such models and raised the need for new approaches. As such, we propose BIAS, a model that, for the first time, combines audio, face, and body information, to accurately predict active speakers in varying/challenging conditions. Additionally, we design BIAS to provide interpretability by proposing a novel use for Squeeze-and-Excitation blocks, namely in attention heatmaps creation and feature importance assessment. For a full interpretability setup, we annotate an ASD-related actions dataset (ASD-Text) to finetune a ViT-GPT2 for text scene description to complement BIAS interpretability. The results show that BIAS is state-of-the-art in challenging conditions where body-based features are of utmost importance (Columbia, open-settings, and WASD), and yields competitive results in AVA-ActiveSpeaker, where face is more influential than body for ASD. BIAS interpretability also shows the features/aspects more relevant towards ASD prediction in varying settings, making it a strong baseline for further developments in interpretable ASD models, and is available at https://github.com/Tiago-Roxo/BIAS.
Timely and accurate assessments of building damage are crucial for effective response and recovery in the aftermath of earthquakes. Conventional preliminary damage assessments (PDA) often rely on manual door-to-door inspections, which are not only time-consuming but also pose significant safety risks. To safely expedite the PDA process, researchers have studied the applicability of satellite imagery processed with heuristic and machine learning approaches. These approaches output binary or, more recently, multiclass damage states at the scale of a block or a single building. However, the current performance of such approaches limits practical applicability. To address this limitation, we introduce a metadata-enriched, transformer based framework that combines high-resolution post-earthquake satellite imagery with building-specific metadata relevant to the seismic performance of the structure. Our model achieves state-of-the-art performance in multiclass post-earthquake damage identification for buildings from the Turkey-Syria earthquake on February 6, 2023. Specifically, we demonstrate that incorporating metadata, such as seismic intensity indicators, soil properties, and SAR damage proxy maps not only enhances the model's accuracy and ability to distinguish between damage classes, but also improves its generalizability across various regions. Furthermore, we conducted a detailed, class-wise analysis of feature importance to understand the model's decision-making across different levels of building damage. This analysis reveals how individual metadata features uniquely contribute to predictions for each damage class. By leveraging both satellite imagery and metadata, our proposed framework enables faster and more accurate damage assessments for precise, multiclass, building-level evaluations that can improve disaster response and accelerate recovery efforts for affected communities.