EMFE: A lightweight, explainable machine learning framework for malaria cell classification
本文提出EMFE框架，通过数学特征提取和经典机器学习方法解决疟疾细胞分类问题，相比深度学习模型更轻量且可解释。
0 citationsRead paper
Selected work
Representative Papers
Discovering Persistent Behavioural Patterns for Interpretable Blockchain Forensics
This study addresses the limitations of existing blockchain behavior analysis methods, which are often confined to specific applications, lack scalability, and offer limited interpretability—hindering long-term, large-scale forensic investigations. To overcome these challenges, this work proposes a scalable, application-agnostic analytical framework that constructs behavioral statements by integrating contract, token, and market context. It models users’ long-term behavioral patterns through sentence-level and sequence-level embeddings and generates interpretable community profiles based on behavioral motifs, routines, and temporal dynamics. Notably, this approach is the first to combine persistent behavioral pattern discovery with explainable analysis, moving beyond conventional snapshot-based methods. Evaluated on 30 million Ethereum transactions, the framework successfully identifies diverse behaviors—including DEX trading, NFT activity, phishing, bot operations, oracle manipulation, and rug-pulls—with most patterns demonstrating stability across time windows.
0 citationsRead paper
VietNormalizer: An Open-Source, Dependency-Free Python Library for Vietnamese Text Normalization in TTS and NLP Applications
This work addresses the challenge posed by prevalent non-standard textual elements in Vietnamese—such as numerals, dates, currencies, abbreviations, and loanwords—which hinder the effective processing of text-to-speech (TTS) and natural language processing (NLP) systems. Existing normalization tools are often either computationally heavy, incompletely covered, or dependent on external services, making standalone deployment impractical. To overcome these limitations, we propose a lightweight, zero-dependency, rule-based unified text normalization pipeline that integrates precompiled regular expressions, a rule engine, CSV-based dictionary mappings, transliteration algorithms, and Unicode normalization. Notably, the system operates without neural networks or external APIs yet comprehensively handles diverse non-standard tokens. The solution is open-sourced via PyPI and GitHub, supports pip installation, and offers high throughput, low memory footprint, and strong extensibility, providing the first self-contained text normalization framework tailored for low-resource tonal languages like Vietnamese.
0 citationsRead paper
A Task-Oriented Evaluation Framework for Text Normalization in Modern NLP Pipelines
Existing stemmer evaluation methods fail to quantify the semantic degradation caused by over-stemming in downstream tasks. This paper proposes the first task-oriented text normalization evaluation framework, overcoming the limitations of traditional lemmatization-based metrics by jointly measuring three dimensions: Stemming Effectiveness Score (SES), Model Performance Delta (MPD), and Average Normalized Levenshtein Distance (ANLD). The framework enables integrated analysis of both efficiency and semantic safety—the first of its kind. Empirical evaluation across multiple languages reveals a critical insight: high stemming recall does not necessarily improve downstream performance. For instance, the Bangla stemmer suffers performance degradation due to aggressive over-stemming, whereas the English Snowball stemmer achieves a superior trade-off between effectiveness and semantic fidelity. This work establishes a principled, task-aware methodology for evaluating and comparing stemmers beyond surface-form matching.
0 citationsRead paper
Concurrent validity of computer-vision artificial intelligence player tracking software using broadcast footage
Commercial AI-based player tracking systems lack systematic validation for concurrent validity in broadcast video, particularly regarding positional accuracy, velocity estimation, and total distance covered during elite football matches. Method: This study conducts the first comprehensive concurrent validity assessment of three leading commercial AI tracking solutions on FIFA World Cup broadcast footage, using multi-camera, high-definition TRACAB Gen 5 data as the ground-truth reference. It quantifies errors in position (RMSE), speed, and cumulative running distance, while analyzing the impact of camera viewpoint (e.g.,俯角, wide-angle) and resolution. Results: Positional RMSE ranges from 1.68–16.39 m; speed error from 0.34–2.38 m/s; and total distance bias from −21.8% to +24.3%. Crucially, tactical camera angles—especially elevated and wide-field views—significantly improve localization accuracy, demonstrating that broadcast geometry fundamentally constrains AI tracking performance. This work establishes the first empirical benchmark and methodological framework for validating AI-driven sports analytics in live broadcast environments.
0 citationsRead paper
Recent publications
Latest Papers
EMFE: A lightweight, explainable machine learning framework for malaria cell classification
本文提出EMFE框架，通过数学特征提取和经典机器学习方法解决疟疾细胞分类问题，相比深度学习模型更轻量且可解释。
0 citationsRead paper
Discovering Persistent Behavioural Patterns for Interpretable Blockchain Forensics
This study addresses the limitations of existing blockchain behavior analysis methods, which are often confined to specific applications, lack scalability, and offer limited interpretability—hindering long-term, large-scale forensic investigations. To overcome these challenges, this work proposes a scalable, application-agnostic analytical framework that constructs behavioral statements by integrating contract, token, and market context. It models users’ long-term behavioral patterns through sentence-level and sequence-level embeddings and generates interpretable community profiles based on behavioral motifs, routines, and temporal dynamics. Notably, this approach is the first to combine persistent behavioral pattern discovery with explainable analysis, moving beyond conventional snapshot-based methods. Evaluated on 30 million Ethereum transactions, the framework successfully identifies diverse behaviors—including DEX trading, NFT activity, phishing, bot operations, oracle manipulation, and rug-pulls—with most patterns demonstrating stability across time windows.
0 citationsRead paper
VietNormalizer: An Open-Source, Dependency-Free Python Library for Vietnamese Text Normalization in TTS and NLP Applications
This work addresses the challenge posed by prevalent non-standard textual elements in Vietnamese—such as numerals, dates, currencies, abbreviations, and loanwords—which hinder the effective processing of text-to-speech (TTS) and natural language processing (NLP) systems. Existing normalization tools are often either computationally heavy, incompletely covered, or dependent on external services, making standalone deployment impractical. To overcome these limitations, we propose a lightweight, zero-dependency, rule-based unified text normalization pipeline that integrates precompiled regular expressions, a rule engine, CSV-based dictionary mappings, transliteration algorithms, and Unicode normalization. Notably, the system operates without neural networks or external APIs yet comprehensively handles diverse non-standard tokens. The solution is open-sourced via PyPI and GitHub, supports pip installation, and offers high throughput, low memory footprint, and strong extensibility, providing the first self-contained text normalization framework tailored for low-resource tonal languages like Vietnamese.
0 citationsRead paper
A Task-Oriented Evaluation Framework for Text Normalization in Modern NLP Pipelines
Existing stemmer evaluation methods fail to quantify the semantic degradation caused by over-stemming in downstream tasks. This paper proposes the first task-oriented text normalization evaluation framework, overcoming the limitations of traditional lemmatization-based metrics by jointly measuring three dimensions: Stemming Effectiveness Score (SES), Model Performance Delta (MPD), and Average Normalized Levenshtein Distance (ANLD). The framework enables integrated analysis of both efficiency and semantic safety—the first of its kind. Empirical evaluation across multiple languages reveals a critical insight: high stemming recall does not necessarily improve downstream performance. For instance, the Bangla stemmer suffers performance degradation due to aggressive over-stemming, whereas the English Snowball stemmer achieves a superior trade-off between effectiveness and semantic fidelity. This work establishes a principled, task-aware methodology for evaluating and comparing stemmers beyond surface-form matching.
0 citationsRead paper
Concurrent validity of computer-vision artificial intelligence player tracking software using broadcast footage
Commercial AI-based player tracking systems lack systematic validation for concurrent validity in broadcast video, particularly regarding positional accuracy, velocity estimation, and total distance covered during elite football matches. Method: This study conducts the first comprehensive concurrent validity assessment of three leading commercial AI tracking solutions on FIFA World Cup broadcast footage, using multi-camera, high-definition TRACAB Gen 5 data as the ground-truth reference. It quantifies errors in position (RMSE), speed, and cumulative running distance, while analyzing the impact of camera viewpoint (e.g.,俯角, wide-angle) and resolution. Results: Positional RMSE ranges from 1.68–16.39 m; speed error from 0.34–2.38 m/s; and total distance bias from −21.8% to +24.3%. Crucially, tactical camera angles—especially elevated and wide-field views—significantly improve localization accuracy, demonstrating that broadcast geometry fundamentally constrains AI tracking performance. This work establishes the first empirical benchmark and methodological framework for validating AI-driven sports analytics in live broadcast environments.
0 citationsRead paper