Yingnan Yi

This study introduces a sophisticated dual-denoising framework for high-frequency momentum strategies in China's ChiNext market, integrating wavelet-based temporal filtering with isolation forest cross-sectional anomaly detection. Utilizing a comprehensive dataset of over 2 million daily observations from January 2016 to November 2025, the results demonstrate that wavelet denoising achieves exceptional efficacy for turnover series with a mean Signal-to-Noise Ratio improvement of 10.7 dB, while isolation forest robustly identifies anomalous stocks characterized by excessive trading activity and distorted risk-return profiles. Empirical results reveal that linear models with wavelet denoising consistently outperform complex alternatives, with Support Vector Machines achieving Sharpe ratios of 0.1132 in long portfolios and 0.0649 in long-short implementations. Contrary to conventional wisdom, simpler architectures demonstrate superior performance, with single-layer neural networks and regularized linear models surpassing deeper architectures across all denoising configurations. The methodology demonstrates remarkable regime invariance, maintaining consistent performance during both low-volatility and high-volatility market states. These findings establish wavelet denoising as a critical preprocessing technique for high-frequency financial data and challenge assumptions about the necessity of complex models in quantitative finance, providing computationally efficient approaches suitable for both institutional and retail investors.

This paper proposes a knowledge-graph-guided generative detection model to address the challenges of complex feature associations, hidden behavioral patterns, and sparse semantic information in financial fraud detection. The study maps multi-source heterogeneous data, such as accounts, transactions, devices, and geographic information, into a knowledge graph structure to model semantic dependencies and behavioral relationships among financial entities at both node and relation levels. Based on this, the model introduces a relation-aware encoder to obtain high-dimensional structural embeddings and employs a generative reasoning mechanism to learn the distribution of latent fraud patterns, enabling semantic identification of complex risk behaviors without relying on fixed rules. During generation, the method integrates knowledge constraints and structural priors, allowing the model to reconstruct potential fraud chains from a global semantic perspective and significantly enhance interpretability and generalization. Validation on real transaction datasets shows that the model outperforms comparison methods in AUC, ACC, Precision, and Recall, demonstrating the effectiveness of the knowledge-enhanced generative framework in capturing cross-entity dependencies and dynamic risk relationships. Overall, the proposed knowledge-graph-guided generative detection model achieves a unified framework of structural modeling, semantic reasoning, and anomaly generation, providing a new systematic solution for the intelligent detection of complex financial fraud behaviors. CCS CONCEPTS: Computing methodologies~Machine learning~Machine learning approaches.

Momentum-based investment strategies face persistent challenges from noise contamination in financial time series, particularly within emerging markets such as China's ChiNext board. Traditional enhancement approaches typically address symptoms rather than underlying causes, resulting in continued vulnerability to market regime changes and performance deterioration. This study develops and evaluates a dual-denoising framework that integrates wavelet analysis for temporal noise reduction with isolation forest algorithms for cross-sectional anomaly detection. Our methodology employs comprehensive analysis of 1,200-1,300 ChiNext stocks spanning the 2015-2025 period, utilizing multiple machine learning architectures to assess portfolio performance across both long-only and long-short implementations. Key empirical findings demonstrate that: denoised momentum factors substantially improve predictive accuracy and portfolio performance; wavelet-based temporal denoising achieves remarkable effectiveness for turnover data with mean signal-to-noise ratio improvements of 6.4 dB; isolation forest cross-sectional anomaly detection provides critical risk management benefits by systematically eliminating stocks characterized by excessive trading activity and poor returns; and single-layer neural networks with isolation forest denoising achieve superior performance metrics, including 0.0199 monthly returns and a 0.2189 Sharpe ratio, outperforming more complex architectural alternatives. Addressing noise contamination at the data level represents a more fundamental solution than conventional enhancement techniques for momentum strategy limitations. Our findings establish systematic denoising as an effective approach for enhancing momentum-based investment strategies while maintaining practical implementability, with significant implications for both quantitative portfolio management and retail investor applications in emerging markets.

This paper addresses the complexity and high-risk nature of fraud detection in financial transaction data by proposing a discrimination method based on an improved Mamba architecture. The study first performs data cleaning, normalization, and embedding to reduce scale differences across multimodal features, and employs temporal encoding and positional embedding to strengthen sequence representation. A dynamic state updating and selective state space mechanism is then introduced to capture long-term dependencies and global contextual features, thereby enhancing the model's discrimination ability in complex transaction environments. Furthermore, context interaction and weighted aggregation modules are designed to integrate hidden states across time steps into a global representation, providing more robust inputs for final classification. Experiments, including baseline comparisons, sequence length sensitivity, and noise robustness tests, demonstrate that the proposed method outperforms multiple mainstream architectures in Accuracy, F1-score, AUC, and Precision, while showing greater stability in handling data imbalance and adversarial perturbations. Overall, the study confirms the effectiveness and applicability of the improved Mamba architecture for financial fraud detection and provides technical support for building secure and trustworthy risk control systems.

This paper addresses key challenges in credit fraud detection, including small sample sizes, imbalanced data distribution, and rapidly changing fraud patterns. A meta-learning-based detection method is proposed to enhance the model's adaptability and detection efficiency in complex scenarios. The method adopts a two-level architecture consisting of a meta-learner and a task-learner. By extracting shared knowledge across multiple fraud detection tasks, the model can rapidly update its parameters. This allows it to maintain high detection performance when facing new types of fraud. In the method design, a task-level gradient update mechanism and a weighted loss strategy are introduced. These components help address the challenges caused by rare fraud samples and class imbalance. The proposed model is systematically evaluated through multiple experiments. These include different task quantities, varying degrees of data imbalance, and injected abnormal fraud patterns. The results confirm the model's effectiveness and robustness under diverse conditions. Experimental findings show that the proposed method outperforms traditional models in accuracy, precision, and recall. It also demonstrates better stability and generalization when facing dynamically changing fraud scenarios. This study highlights the application potential of meta-learning in credit fraud detection. It provides technical support for building more intelligent financial risk control models.