METARO^(3):Metamorphic Relation Group for Automatic Program Repair

The application of metamorphic testing(MT)on automatic program repair(APR-MT)is used to generate a patch without test oracles by examining whether the input metamorphic relation(MR)is satisfied or not.However,the delivered patch is plausible since it may satisfy the input MR but violate other MRs.This inspires us to propose an improved approach to enhance the effectiveness of APR-MT with metamorphic relation group.Ourapproach involves three major steps.First,we formally define the repair process of APR-MT by building the model of automatic program repair and metamorphic testing separately.Then,we propose the advanced model of automatic program repair based on metamorphic relation group,named METARO^(3),which takes several MRs as input while only one MR is used in APR-MT.We additionally present two kinds of selection strategies to rank MRs in descending order of the fault detection capability,which helps shorten the repair time of finding a patch.To demonstrate the feasibility and procedure of our approach,an illustration example was conducted.The results show that METARO^(3) can improve the effectiveness of APR-MT significantly.

Multi-View Feature Fusion Model for Software Bug Repair Pattern Prediction

Many search-based Automatic Program Repair(APR)techniques employ a set of repair patterns to generate candidate patches.Regarding repair pattern selection,existing search-based APR techniques either randomly select a repair pattern from the repair pattern set to apply or prioritize all repair patterns based on the bug's context information.In this paper,we introduce PatternNet,a multi-view feature f usion model capable of predicting the repair pattern for a reported software bug.To accomplish this task,PatternNet first extracts multiview features from the pair of buggy code and bug report using different models.Specifically,a transformer-based model(i.e.,UniXcoder)is utilized to obtain the bimodal feature representation of the buggy code and bug report.Additionally,an Abstract Syntax Tree(AST)-based neural model(i.e.,ASTNN)is employed to learn the feature representation of the buggy code.Second,a co-attention mechanism is adopted to capture the dependencies between the statement trees in the AST of the buggy code and the textual tokens of the reported bug,resulting in co-attentive features between statement trees and reported bug's textual tokens.Finally,these multi-view features are combined i nto a unified representation using a feature fusion network.We quantitatively demonstrate the effectiveness of PatternNet and the feature fusion network for predicting software bug repair patterns.

Predicting Effectiveness of Generate-and-Validate Patch Generation Systems Using Random Forest

One way to improve practicability of automatic program repair（APR） techniques is to build prediction models which can predict whether an application of a APR technique on a bug is effective or not. Existing prediction models have some limitations. First, the prediction models are built with hand crafted features which usually fail to capture the semantic characteristics of program repair task. Second, the performance of the prediction models is only evaluated on Genprog, a genetic-programming based APR technique. This paper develops prediction models, i.e., random forest prediction models for SPR, another kind of generate-and-validate APR technique, which can distinguish ineffective repair instances from effective repair instances. Rather than handcrafted features, we use features automatically learned by deep belief network（DBN） to train the prediction models. The empirical results show that compared to the baseline models, that is, all effective models, our proposed models can at least improve the F1 by 9% and AUC（area under the receiver operating characteristics curve） by 19%. At the same time, the prediction model using learned features at least outperforms the one using hand-crafted features in terms of F1 by 11%.