pattern matching

Students commonly struggle with constructing and debugging regular expressions due to cognitive overload and lack of systematic guidance. Method: This paper proposes a pedagogically oriented, structured design support framework that integrates the “design recipe” methodology to decompose regex construction into incremental, actionable steps; introduces a novel, stage-based error classification model that generates concise, jargon-free, non-prescriptive feedback; and incorporates a lightweight unit-testing shorthand syntax to lower verification barriers. Contribution/Results: Classroom deployment demonstrates significant improvements in students’ correctness rates and conceptual understanding during regex construction. Two authentic debugging case studies confirm the framework’s usability and pedagogical effectiveness, validating its capacity to enhance both learning efficiency and diagnostic accuracy in regex education.

This study addresses the challenge of automatically detecting software design patterns in source code to support architectural understanding and quality assessment. It presents the first systematic evaluation of four large language models—including NextCoder and Gemma 3—as well as two ensemble strategies combining three models, for recognizing five classic design patterns: Singleton, Adapter, Bridge, Composite, and Decorator. The work investigates the impact of three input modalities—raw source code, PlantUML diagrams, and textual descriptions—on detection performance. Experimental results demonstrate that NextCoder and Gemma 3 achieve the highest accuracy among individual models, while ensemble approaches further enhance performance, thereby confirming the effectiveness and potential of large language models in design pattern recognition tasks.

This paper addresses periodicity detection in noisy string streams, supporting realistic perturbations including mismatches, wildcards, and edit operations. For Hamming distance, we present the first single-pass streaming algorithm for period detection that imposes no restrictions on suffix characters and fully supports wildcards; for edit distance, we design the first two-pass streaming algorithm. Our approach integrates and significantly extends three key techniques: Clifford et al.’s Hamming sketch, Charalampopoulos’ structural analysis for k-mismatch periodicity, and Bhattacharya–Koucký’s grammar-based decomposition. Compared to prior work, our algorithms achieve improved time complexity, eliminate traditional constraints on wildcard positions, and demonstrate superior robustness and practicality on real-world noisy data.

This work addresses the challenge of efficiently representing and uniformly operating on large-scale collections of regular languages. Methodologically, it introduces a hierarchical normalization framework that integrates syntactic derivative theory with equivalence-class partitioning, yielding a two-tiered structure: an upper tier normalizes regular expressions syntactically, while a lower tier represents each language by its minimal deterministic finite automaton (DFA) and assigns it a unique integer identifier—ensuring a finite, computable, and canonical set of derivatives per language. The key contribution is the first principled unification of expression normalization, language-equivalence compression, and integer encoding, enabling compact and canonical representation of regular languages. Experiments demonstrate substantial improvements in storage efficiency and operational speed for large language collections, supporting fast equivalence checking, Boolean operations (intersection, union, complement), and membership queries—balancing theoretical rigor with practical deployability.

The absence of automated, systematic approaches for modeling pattern recognition in conceptual modeling hinders improvements in knowledge representation and modeling quality. Method: This paper introduces frequent subgraph mining—systematically applied to conceptual modeling for the first time—and proposes a cross-language (OntoUML/ArchiMate), multi-criteria structural pattern discovery framework. It integrates a gSpan variant with graph editing, graph isomorphism testing, and pattern abstraction techniques to build an extensible, exploratory analysis tool. Contribution/Results: Evaluated on two authoritative datasets, the method successfully identifies highly reusable structural patterns. It demonstrates effectiveness in assessing modeling practices, supporting language evolution, and optimizing model quality—thereby filling a critical research gap in automated pattern mining for conceptual modeling.

This work addresses the computational bottleneck of exhaustive search in interval pattern mining under user-defined syntactic constraints. To overcome this limitation, the authors propose CFips, a novel method that, for the first time, supports diverse syntactic constraints while guaranteeing exact sampling in interval pattern mining. CFips employs a multi-step sampling framework that decomposes complex constraints into elementary predicates over interval boundaries and integrates a frequency-weighted strategy to achieve efficient sampling strictly aligned with the true distribution within the constrained space. Theoretical analysis establishes the correctness of the sampling procedure, and empirical results demonstrate that CFips successfully completes tasks within practical time limits that are intractable for conventional approaches, substantially enhancing both the efficiency and feasibility of constrained pattern mining.

This study addresses the lack of empirical validation regarding the practical application of AI design patterns in real-world codebases. By mining the literature, the authors identify 14 distinct AI design patterns and, for the first time, integrate active learning with pattern mining to construct a quantifiable analytical framework across 100 open-source AI projects on GitHub. This framework enables estimation of the occurrence frequency and statistical confidence bounds for each pattern. Evaluated on an 8-class classification task, the approach achieves 56% accuracy and 55% recall—substantially outperforming the 11% random baseline. The work fills a critical gap in empirical research on AI design patterns and provides practitioners with actionable estimates of pattern applicability boundaries.

This study addresses the lack of systematic comparison between deterministic (heuristic-based) and probabilistic (large language model–based) approaches for generating Java code summaries focused on design patterns. Conducting a controlled empirical evaluation on a structured corpus of design pattern–oriented code, we compare three methods—rule-based NLG, SWUM, and Mixtral—along dimensions of semantic alignment, contextual awareness, and conciseness. Evaluation employs BERTScore, cosine similarity, and Llama 3–generated scoring rubrics, complemented by Wilcoxon signed-rank tests, Friedman tests, and Spearman correlation analyses for multidimensional quantification. Results reveal that probabilistic methods excel in semantic alignment and contextual coverage, whereas deterministic approaches yield more concise and fully reproducible outputs, highlighting a clear trade-off between accuracy, brevity, and stability.

This study addresses the prevalent yet underexplored issue of refactorable repetitive step subsequences in Behavior-Driven Development (BDD) tests, for which no automated identification and classification methods previously existed. We propose the first end-to-end framework that leverages Sentence-BERT, UMAP, and HDBSCAN to perform semantic clustering on Gherkin corpora, thereby uncovering recurring fragments. These fragments are then annotated manually to train an XGBoost classifier that ranks their refactoring potential and assigns them to one of three established refactoring patterns. Applying our approach across 339 repositories, we identify 692,020 repetitive patterns and release the first large-scale annotated dataset, along with a complete toolchain and evaluation benchmark. Experimental results demonstrate that our XGBoost classifier achieves an F1 score of 0.891, significantly outperforming rule-based baselines and LLM-based judges, with 75% of test scenarios containing high-potential refactoring candidates.

This work proposes a language- and syntax-agnostic approach to string similarity measurement by introducing co-occurrence matrices (COM) and run-length matrices (RLM)—concepts originally from image texture analysis—into the domain of string representation to construct purely statistical, language-independent features. The method integrates multiple statistical measures, including COM, RLM, longest common subsequence, and edit distance. Evaluated on synthetic datasets, COM and RLM significantly outperformed baseline methods in three out of four experiments (p < 0.001). In real-world text plagiarism detection tasks, RLM achieved the best performance, demonstrating the effectiveness and generalizability of the proposed statistical features.