An Effective Semantic Code Clone Detection Framework Using Pairwise Feature Fusion

Abstract

Code clone detection is important for effective software maintenance. The task is more challenging when clones are semantically similar (Type-IV) in nature, having no structural resemblance to each other. Most existing methods use sequence similarity and/or graph isomorphism between either Abstract Syntax Trees (AST) or Program Dependency Graphs (PDG) to detect Type-I, II and III clones. However, they are mostly unsuccessful in detecting semantic or Type-IV clones. In this work, we propose a novel detection framework using machine learning for automated detection of all four type of clones. The features extracted from a pair code blocks are combined for possible detection of a clone with respect to reference block. We use ASTs and PDGs features of both the code blocks to prepare labelled training samples after fusing the two feature vectors using three different alternatives. We use six (06) state-of-the-art classification models including Deep Convolutional Neural Network to assess the prediction performance of our scheme. To access the effectiveness of our framework we use seven (07) datasets and compare its performance with five state-of-the-art clone detectors. We also compare a large number of algorithms for code clone detection. Comparing the performance of a large number of machine learning techniques, ANN and non-ANN, using such features, and establishing that fusing of AST and PDG features gives competitive results using deep learning as well as boosted tree algorithms, and find that boosted tree algorithms like XGBOOT are quite competitive in clone detection. Experimental results demonstrate that our approach outperforms existing clone detection methods in terms of prediction accuracy.

A novel code representation for detecting Java code clones using high-level and abstract compiled code representations

“…It achieves higher recall and F1 scores, with values of (96.3%, 89.7%) and (96.4%, 92.3%) respectively. Furthermore, the proposed technique, utilizing the LightBGM classifier outperformed CDLH, ASTNN, and Sheneamer et al [78] in detecting VST3, ST3, and MT3 clones in terms of F1-score.…”

“…ASTNN [77] employs an AST-based neural network that represents source code with smaller statement trees derived from segmented ASTs, capturing both lexical and syntactical knowledge. Finally, Sheneamer et al [78] employ multiple machine learning classifiers to detect code clones, using features extracted from ASTs and PDGs. DLC [76] uses a recursive neural network approach to detect code similarity, utilizing the Euclidean distance as a metric.…”

“…DLC [76] uses a recursive neural network approach to detect code similarity, utilizing the Euclidean distance as a metric. [7,28,31,45,[75][76][77][78].…”

A Collaborative method for Code Clone Detection Using Lexical, Syntactic, Semantic and Structural Features

“…An efficient semantic CCD (ES-CCD) technique was developed [13] by employing a pair-wise feature fusion for automated identification of all four clone types. AST and program dependency graphs (PDGs) were efficiently utilized to prepare labelled training features for detecting the Java code clones, including semantic clones.…”

“…Appache opennlp-master 1.9.1 is a natural language information retrieval toolkit based on machine learning. Among the most common NLP operations supported are indexing, sentence classification, part-of-speech labelling, named entity identification, stacking, and processing [25] Using these datasets, the proposed method's efficiency is compared to that of existing methods such as LV-CCD [9], ES-CCD [13], TBCNN-CCD [14], and CPVDetector [15] terms of accuracy, precision, recall, time period, memory space, and clone types for detecting all four types of CC detection and its similarity features.…”