Analyzing Stack Flows to Compare Java Programs

Abstract

SUMMARYThis paper presents a method for comparing and detecting clones of Java programs by analyzing program stack flows. A stack flow denotes an operational behavior of a program by describing individual instructions and stack movements for performing specific operations. We analyze stack flows by simulating the operand stack movements during execution of a Java program. Two programs for detection of clones of Java programs are compared by matching similar pairs of stack flows in the programs. Experiments were performed on the proposed method and compared with the earlier approaches of comparing Java programs, the Tamada, k-gram, and stack pattern based methods. Their performance was evaluated with real-world Java programs in several categories collected from the Internet. The experimental results show that the proposed method is more effective than earlier methods of comparing and detecting clones of Java programs.

Revisiting the Challenges and Opportunities in Software Plagiarism Detection

“…For R2, most dynamic analysis binary based approaches, (e.g., DYKIS [30], TOB [28], TreSB [27], and LoPD [21]) would fail because they detect whole program plagiarism and only generate a similarity score. The approaches based on static analysis (e.g., SKB [22], SWKB [33], SFB [17] and WSPB [18]) can satisfy R2, because static analysis make it possible to be applied at different granularities, such as basic blocks, functions, which can be detected locally.…”

“…AST [35] based on abstract syntax tree only partially satisfy R3, as a result of failing to handle advanced code obfuscation, for example, instruction rearrangement, statement splitting. SFB [17] and WSPB [18] also only partially satisfy R3, because they are vulnerable to control flow obfuscation. GPLAG [19] which relies the analysis on Program Dependency Graphs (PDGs) can perform better robustness against advanced semantics-preserving code obfuscation techniques and satisfies R3.…”

Software Plagiarism Detection with Birthmarks Based on Dynamic Key Instruction Sequences

“…Decision maker decides plagiarism by the average value ψ of multiple similarity scores against a predefined threshold ε. We adopt a default value of ε = 0.25 as in previous studies [11], [16], [31], [32]. That is, two programs are classified as independent if ψ ≤ 0.25, as plagiarized if ψ ≥ 0.75.…”

“…Lim [34] used control flow information that reflected runtime behaviors to supplement static approaches. Recently he proposed to analyze stack flows obtained by simulating operand stack movements to detect copies [31]. Yet they are only available to Java programs.…”

Plagiarism detection for multithreaded software based on thread-aware software birthmarks

“…Lim proposed to use control flow information that reflects runtime behaviors to supplement static approaches [13]. More recently Lim proposed to analyze stack flows obtained by simulating the operand stack movements to detect copies [12].…”