Search

In response to the limitations of traditional fuzzing approaches that rely on static mutators and fail to dynamically adjust their test case mutations for deeper testing, resulting in the inability to generate targeted inputs to trigger vulnerabilities, this paper proposes a directed fuzzing methodology termed DocFuzz, which is predicated on a feedback mechanism mutator. Initially, a sanitizer is used to target the source code of the tested program and stake in code blocks that may have vulnerabilities. After this, a taint tracking module is used to associate the target code block with the bytes in the test case, forming a high‐value byte set. Then, the reinforcement learning mutator of DocFuzz is used to mutate the high‐value byte set, generating well‐structured inputs that can cover the target code blocks. Finally, utilizing the feedback mechanism of DocFuzz, when the reinforcement learning mutator converges and ceases to optimize, the fuzzer is rebooted to continue mutating toward directions that are more likely to trigger vulnerabilities. Comparative experiments are conducted on multiple test sets, including LAVA‐M, and the experimental results demonstrate that the proposed DocFuzz methodology surpasses other fuzzing techniques, offering a more precise, rapid, and effective means of detecting vulnerabilities in source code. [ABSTRACT FROM AUTHOR]

Metamorphic testing (MT) is an effective testing technique having a broad range of applications. One key task for MT is the identification of metamorphic relations (MRs), which is a fundamental mechanism in MT and is critical to the automation of MT. Prior studies have proposed approaches for predicting MRs (PMR). One major idea behind these PMR approaches is to represent program source code information via manually designed code features and then to apply machine‐learning–based classifiers to automatically predict whether a specific MR can be applied on the target program. Nevertheless, the human‐involved procedure of selecting and extracting code features is costly, and it may not be easy to obtain sufficiently comprehensive features for representing source code. To overcome this limitation, in this study, we explore and evaluate the effectiveness of code representation learning techniques for PMR. By applying neural code representation models for automatically mapping program source code to code vectors, the PMR procedure can be boosted with learned code representations. We develop 32 PMR instances by, respectively, combining 8 code representation models with 4 typical classification models and conduct an extensive empirical study to investigate the effectiveness of code representation learning techniques in the context of MR prediction. Our findings reveal that code representation learning can positively contribute to the prediction of MRs and provide insights into the practical usage of code representation models in the context of MR prediction. Our findings could help researchers and practitioners to gain a deeper understanding of the strength of code representation learning for PMR and, hence, pave the way for future research in deriving or extracting MRs from program source code. [ABSTRACT FROM AUTHOR]

Code clone detection is a critical problem in software development and maintenance domains. It aims to identify functionally identical or similar code fragments within an application. Existing works formulate the code clone detection task as a binary classification problem which predicts a code pair as a clone or not based on a pre‐defined threshold. In reality, there are various types of code clone subject to the degree of how a pair of code fragments are similar to each other. To investigate the effect of different code clone detection manners on the clone detection result, we propose Gated Recurrent Residual Learning Networks (GRRLN), a novel neural network model for code clone detection. To train GRRLN, we first represent each code fragment as a statement‐level tree sequence derived from the whole abstract syntax tree (AST). Then, a gated recurrent neural network with residual connections is adopted to fully extract the semantics of all individual statement trees together with their dependency relationships across the input statement sequence. Finally, the output representations of code fragments by GRRLN are used for similarity calculation and clone detection. We evaluate GRRLN using two real‐world datasets for code clone detection and clone type classification. Experiments show that GRRLN achieves promising and compelling results and meanwhile needs significantly less time and memory consumption compared with the state‐of‐the‐art methods. [ABSTRACT FROM AUTHOR]

Software quality is key to the success of software systems. Modern software systems are growing in their worth based on industry needs and becoming more complex, which inevitably increases the possibility of more defects in software systems. Software repairing is time‐consuming, especially locating the source files related to specific software defect reports. To locate defective source files more quickly and accurately, automated software defect location technology is generated and has a huge application value. The existing deep learning‐based software defect location method focuses on extracting the semantic correlation between the source file and the corresponding defect reports. However, the extensive code structure information contained in the source files is ignored. To this end, we propose a software defect location method, namely, multi‐graph learning‐based software defect location (MGSDL). By extracting the program dependency graphs for functions, each source file is converted into a graph bag containing multiple graphs (i.e., multi‐graph). Further, a multi‐graph learning method is proposed, which learns code structure information from multi‐graph to establish the semantic association between source files and software defect reports. Experiments' results on four publicly available datasets, AspectJ, Tomcat, Eclipse UI, and SWT, show that MGSDL improves on average 3.88%, 5.66%, 13.23%, 9.47%, and 3.26% over the competitive methods in five evaluation metrics, rank@10, rank@5, MRR, MAP, and AUC, respectively. [ABSTRACT FROM AUTHOR]

Over the past few years, the software engineering (SE) community has widely employed deep learning (DL) techniques in many source code processing tasks. Similar to other domains like computer vision and natural language processing (NLP), the state‐of‐the‐art DL techniques for source code processing can still suffer from adversarial vulnerability, where minor code perturbations can mislead a DL model's inference. Efficiently detecting such vulnerability to expose the risks at an early stage is an essential step and of great importance for further enhancement. This paper proposes a novel black‐box effective and high‐quality adversarial attack method, namely CodeBERT‐Attack (CBA), based on the powerful large pre‐trained model (i.e., CodeBERT) for DL models of source code processing. CBA locates the vulnerable positions through masking and leverages the power of CodeBERT to generate textual preserving perturbations. We turn CodeBERT against DL models and further fine‐tuned CodeBERT models for specific downstream tasks, and successfully mislead these victim models to erroneous outputs. In addition, taking the power of CodeBERT, CBA is capable of effectively generating adversarial examples that are less perceptible to programmers. Our in‐depth evaluation on two typical source code classification tasks (i.e., functionality classification and code clone detection) against the most widely adopted LSTM and the powerful fine‐tuned CodeBERT models demonstrate the advantages of our proposed technique in terms of both effectiveness and efficiency. Furthermore, our results also show (1) that pre‐training may help CodeBERT gain resilience against perturbations further, and (2) certain pre‐training tasks may be beneficial for adversarial robustness. [ABSTRACT FROM AUTHOR]

Summary: Machine learning‐based code smell detection (CSD) has been demonstrated to be a valuable approach for improving software quality and enabling developers to identify problematic patterns in code. However, previous researches have shown that the code smell datasets commonly used to train these models are heavily imbalanced. While some recent studies have explored the use of imbalanced learning techniques for CSD, they have only evaluated a limited number of techniques and thus their conclusions about the most effective methods may be biased and inconclusive. To thoroughly evaluate the effect of imbalanced learning techniques for machine learning‐based CSD, we examine 31 imbalanced learning techniques with seven classifiers to build CSD models on four code smell data sets. We employ four evaluation metrics to assess the detection performance with the Wilcoxon signed‐rank test and Cliff's δ$$ \delta $$. The results show that (1) Not all imbalanced learning techniques significantly improve detection performance, but deep forest significantly outperforms the other techniques on all code smell data sets. (2) SMOTE (Synthetic Minority Over‐sampling TEchnique) is not the most effective technique for resampling code smell data sets. (3) The best‐performing imbalanced learning techniques and the top‐3 data resampling techniques have little time cost for code smell detection. Therefore, we provide some practical guidelines. First, researchers and practitioners should select the appropriate imbalanced learning techniques (e.g., deep forest) to ameliorate the class imbalance problem. In contrast, the blind application of imbalanced learning techniques could be harmful. Then, better data resampling techniques than SMOTE should be selected to preprocess the code smell data sets. [ABSTRACT FROM AUTHOR]

Background: The accurate encoding of operation notes is essential for activity‐based funding and workforce planning. The aim of this project was to evaluate the procedural coding accuracy of vitrectomy and to develop machine learning, natural language processing (NLP) models that may assist with this task. Methods: This retrospective cohort study involved vitrectomy operation notes between a 21‐month period at the Royal Adelaide Hospital. Coding of procedures were based on the Medicare Benefits Schedule (MBS)—the Australian equivalent to the Current Procedural Terminology (CPT®) codes used in the United States. Manual encoding was conducted for all procedures and reviewed by two vitreoretinal consultants. XGBoost, random forest and logistic regression models were developed for classification experiments. A cost‐based analysis was subsequently conducted. Results: There were a total of 1724 procedures with individual codes performed within 617 vitrectomy operation notes totalling $1 528 086.60 after manual review. A total of 1147 (66.5%) codes were missed in the original coding that amounted to $736 539.20 (48.2%). Our XGBoost model had the highest classification accuracy (94.6%) in the multi‐label classification for the five most common procedures. The XGBoost model was the most successful model in identifying operation notes with two or more missing codes with an AUC of 0.87 (95% CI 0.80–0.92). Conclusions: Machine learning has been successful in the classification of vitrectomy operation note encoding. We recommend a combined human and machine learning approach to clinical coding as automation may facilitate more accurate reimbursement and enable surgeons to prioritise higher quality clinical care. [ABSTRACT FROM AUTHOR]

Summary: With the wide application of deep learning in code search, especially the proposed code search model based on attention mechanism, the accuracy of code search has been greatly improved. However, the attention mechanism only captures the attention weight relationship between two words in the code fragment, without considering the contextual semantic relationship that exists between words in the code fragment, which can help improve the accuracy of code search. To address this problem, this paper proposes a model that combining bidirectional long short‐term memory and self‐attention mechanisms for code search (CBLSAM‐CS). The model first captures the contextual semantic relationship of each word in the code fragment by long‐short term memory network, and then uses the self‐attention mechanism to extract deep‐level features of the sequence. In order to verify the effectiveness of the proposed model, the paper has been conducted an experimental comparison with three other baseline models, CODEnn, CARLCS‐CNN, and SAN‐CS, on the basis of a public dataset containing 18 million code fragments. The experimental results show that the proposed model in this paper achieves 92.24% and 93.55% in mean reciprocal rank value and normalized discounted cumulative gain metrics, respectively, which are better than the baseline model. Therefore, it shows that the CBLSAM‐CS model proposed in this paper can effectively improve the accuracy and efficiency of code search. [ABSTRACT FROM AUTHOR]

The rapid development of Open‐Source Software (OSS) has resulted in a significant demand for code changes to maintain OSS. Symptoms of poor design and implementation choices in code changes often occur, thus heavily hindering code reviewers to verify correctness and soundness of code changes. Researchers have investigated how to learn meaningful code changes to assist developers in anticipating changes that code reviewers may suggest for the submitted code. However, there are two main limitations to be addressed, including the limitation of long‐range dependencies of the source code and the missing syntactic structural information of the source code. To solve these limitations, a novel method is proposed, named Graph Transformer for learning meaningful Code Transformations (GTCT), to provide developers with preliminary and quick feedback when developers submit code changes, which can improve the quality of code changes and improve the efficiency of code review. GTCT comprises two components: code graph embedding and code transformation learning. To address the missing syntactic structural information of the source code limitation, the code graph embedding component captures the types and patterns of code changes by encoding the source code into a code graph structure from the lexical and syntactic representations of the source code. Subsequently, the code transformation learning component uses the multi‐head attention mechanism and positional encoding mechanism to address the long‐range dependencies limitation. Extensive experiments are conducted to evaluate the performance of GTCT by both quantitative and qualitative analyses. For the quantitative analysis, GTCT relatively outperforms the baseline on six datasets by 210%, 342.86%, 135%, 29.41%, 109.09%, and 91.67% in terms of perfect prediction. Meanwhile, the qualitative analysis shows that each type of code change by GTCT outperforms that of the baseline method in terms of bug fixed, refactoring code and others' taxonomy of code changes. [ABSTRACT FROM AUTHOR]

Drawing on recent research on robust governance, we conceptualize robust crisis communication as a dynamic process centered on evolving public communication demands. We propose a three‐dimensional measurement for empirically examining the robustness of government crisis communication in the context of the COVID‐19 pandemic. We collected 43,642 Twitter messages posted by 50 state governors in the United States from January 1 to June 30, 2020. We applied machine learning algorithms to code the voluminous Twitter data based on messaging topics, sentiments, and interactions. This study found an overall low level of robustness in the governors' crisis communication. Governors most frequently posted reputation management tweets, followed by tweets about the government's handling of the pandemic. This research presents empirical evidence for the heavy influence of politics on governors' crisis communication strategies and highlights the need to understand and build robust crisis communication. 摘要 本文借鉴稳健性治理的最新相关研究，提出稳健性危机沟通的概念，将其界定为聚焦于满足危机中持续变化的公众沟通需求的动态沟通过程与模式。基于此概念，本文以新型冠状病毒肺炎疫情中的危机沟通为例，构建三维测量方法，用以检验政府危机沟通在内容主题、情感表达和对话互动三方面的稳健程度。本文收集美国50个州长在2020年1月1日至2020年6月30日期间发布的43,642条推特信息，并应用机器学习对所有信息的主题、情感和互动特征进行编码分析。本文发现美国州长危机沟通的稳健性程度总体偏低。美国州长最经常发布与声誉管理相关的推文，其次是政府的疫情防控措施。文章进一步阐释与例证政治因素对州长危机沟通策略的深刻影响，并强调政府部门理解与践行稳健性风险沟通的必要性与重要性。 [ABSTRACT FROM AUTHOR]

Deep learning based image quality assessment models usually learn to predict image quality from a single dataset, leading the model to overfit specific scenes. To account for this, mixed datasets training can be an effective way to enhance the generalization capability of the model. However, it is nontrivial to combine different image quality assessment datasets, as their quality evaluation criteria, score ranges, view conditions, as well as subjects are usually not shared during the image quality annotation. Instead of aligning the annotations, this paper proposes a monotonic neural network for image quality assessment model learning with different datasets combined. In particular, this model consists of a dataset‐shared quality regressor and several dataset‐specific quality transformers. The quality regressor aims to obtain the perceptual quality of each image of each dataset and the quality transformer maps the perceptual quality to the corresponding annotation monotonically. The experimental results verify the effectiveness of the proposed learning strategy and the code is available at https://github.com/fzp0424/MonotonicIQA. [ABSTRACT FROM AUTHOR]

In this paper, the application of instant messaging (IM) in programming training is argued to be possible and inevitable, especially in the high schools/colleges of poor communities. However, challenges in its application were noted. These challenges of utilising IM and the reluctance of the students in applying it in their programming training acted as a motivation for the study. Our aim is to use IM for a different purpose than its original intention. Critical systems heuristics was used as a multi‐methodological framework for the incorporation of various conditioned realities, where the technological pedagogical content knowledge framework was applied to incorporate the applicable technological, pedagogical, content areas. The guidelines were enhanced from knowledge gained from an action research project with trainee teachers. This iterative process demonstrates the incorporation of both theoretical and empirical context‐based perspectives into the understanding and addressing of a problem environment. [ABSTRACT FROM AUTHOR]

Currently, software defect-prediction technology is being extensively researched in the design of metrics. However, the research objects are mainly limited to coarse-grained entities such as classes, files, and packages, and there is a wide range of defects that are difficult to predict in actual situations. To further explore the information between sequences of method calls and to learn the code semantics and syntactic structure between methods, we generated a method-call sequence that retains the code context structure information and the token sequence representing semantic information. We embedded the token sequence into the method-call sequence and encoded it into a fixed-length real-valued vector. We then built a defect-prediction model based on the transformer, which maps the code-vector representation containing the method-call sequences to a low-dimensional vector space to generate semantic features and syntactic structure features and also predicts the defect density of the method-call sequence. We conducted experiments on 10 open-source projects using the ELFF dataset. The experimental results show that the method-call sequence-level prediction effect is better than the class-level effect, and the prediction results are more stable than those of the method level. The mean absolute error (MAE) value of our approach was 8% lower than that of the other deep-learning methods. [ABSTRACT FROM AUTHOR]

Using multi‐task learning to extract code features can effectively increase the information of the features. However, the existing multi‐task learning methods mainly have two limitations: (1) They cannot extract enough code‐related information or only extract similar semantic features. Similar multi‐task makes the information in the features increased insufficiently. However, the high difference multi‐task is challenging to converge. (2) They cannot train multi‐task on heterogeneous datasets. In standard multi‐task training, we need to label all tasks for all data, which consumes enormous labor. To solve the above limitations, we select two high difference tasks, the cross‐language code completion task and variable misuse task, to extract expressive semantic code features. We propose an attention‐based feature fusion module to merge information among high difference tasks, avoiding the convergence dilemma of standard multi‐task learning. We propose a federated learning framework, extracting semantic information and using the feature fusion module to integrate multi‐task information among single labeled datasets. We experiment on C# and Python datasets for the code completion and variable misuse tasks. The results show that the performance of fusion features by FedMTFF improved by up to 22.6% and 15.1% compared to single tasks. We use FedMTFF to perform four cross‐language multi‐task features fusion, exceeding the current best baseline by 24.1%. [ABSTRACT FROM AUTHOR]

The participatory view of learning emphasises students' identity construction. However, identity research in the context of programming education to cultivate students' computational thinking is scarce. In this study, an instrument of computational identity with components of engagement, imagination and affiliation, was developed and validated. Convenience sampling was used to select 1066 senior primary school students studying programming to respond to the instrument, and to a programming empowerment instrument developed previously with components of meaningfulness, self‐efficacy and impact. The two instruments' factor structure was confirmed to have acceptable discriminant validity and support gender invariance. There was a positive relationship between programming empowerment and computational identity. Specifically, students' perceived meaningfulness and self‐efficacy of programming related to all aspects of computational identity. The perceived impact of programming is related only to the imaginative aspect of computational identity. The establishment of the instrument enables researchers to investigate further factors related to students' computational identity development. The results also indicate that the programming curriculum should be carefully designed so that students can realise the meaning of the activities and foster their programming self‐efficacy. This, in turn, is critical to enable these primary school students to participate in and become a member of the digital community. Practitioner notesWhat is already known about this topic Cultivating students' computational thinking (CT) has become an educational goal in various countries around the world.It is necessary to develop appropriate tools to assess students' CT development.Founded on a participatory view of learning, identity has been employed in different subject areas to assess students' development, but identity research in CT is scarce.What this paper adds An instrument of computational identity with components of engagement, imagination and affiliation is developed and validated with a sample of senior primary students.This study shows the importance of empowerment in developing primary students' computational identity.Students' perceived meaningfulness of programming and self‐efficacy relate to all aspects of identity, but the impact of programming only relates to the imaginative aspect of identity.Implications for practice and/or policy The curriculum for CT should be designed carefully to foster the development of students' computational identity.The curriculum should allow students to see the meaning of programming activities and how young people can make an impact by means of CT.The programming tasks should be designed at an optimal level of difficulty so that students' self‐efficacy can be fostered. [ABSTRACT FROM AUTHOR]

Synapses play a critical role in memory, learning, and cognition. Their main functions include converting presynaptic voltage spikes to postsynaptic currents, as well as scaling the input signal. Several brain‐inspired architectures have been proposed to emulate the behavior of biological synapses. While these are useful to explore the properties of nervous systems, the challenge of making biocompatible and flexible circuits with biologically plausible time constants and tunable gain remains. Here, a physically flexible organic log‐domain integrator synaptic circuit is shown to address this challenge. In particular, the circuit is fabricated using organic‐based materials that are electrically active, offer flexibility and biocompatibility, as well as time constants (critical in learning neural codes and encoding spatiotemporal patterns) that are biologically plausible. Using a 10 nF synaptic capacitor, the time constant reached 126 and 221 ms before and during bending, respectively. The flexible synaptic circuit is characterized before and during bending, followed with studies on the effects of weighting voltage, synaptic capacitance, and disparity in presynaptic signals on the time constant. [ABSTRACT FROM AUTHOR]

Due to multitudinous vulnerabilities in sophisticated software programs, the detection performance of existing approaches requires further improvement. Multiple vulnerability detection approaches have been proposed to aid code inspection. Among them, there is a line of approaches that apply deep learning (DL) techniques and achieve promising results. This paper attempts to utilize CodeBERT which is a deep contextualized model as an embedding solution to facilitate the detection of vulnerabilities in C open-source projects. The application of CodeBERT for code analysis allows the rich and latent patterns within software code to be revealed, having the potential to facilitate various downstream tasks such as the detection of software vulnerability. CodeBERT inherits the architecture of BERT, providing a stacked encoder of transformer in a bidirectional structure. This facilitates the learning of vulnerable code patterns which requires long-range dependency analysis. Additionally, the multihead attention mechanism of transformer enables multiple key variables of a data flow to be focused, which is crucial for analyzing and tracing potentially vulnerable data flaws, eventually, resulting in optimized detection performance. To evaluate the effectiveness of the proposed CodeBERT-based embedding solution, four mainstream-embedding methods are compared for generating software code embeddings, including Word2Vec, GloVe, and FastText. Experimental results show that CodeBERT-based embedding outperforms other embedding models on the downstream vulnerability detection tasks. To further boost performance, we proposed to include synthetic vulnerable functions and perform synthetic and real-world data fine tuning to facilitate the model learning of C-related vulnerable code patterns. Meanwhile, we explored the suitable configuration of CodeBERT. The evaluation results show that the model with new parameters outperform some state-of-the-art detection methods in our dataset. [ABSTRACT FROM AUTHOR]

Applying an external electric AC field to a dusty plasma, the micro‐particles arrange in strings or chains. In analogy to electrorheological fluids, such a system is called electrorheological plasma. Turning gradually the AC field into a DC field, the string formation is diminished until it vanishes completely in the DC case. In this way, a cross‐over transition from a string‐like to an isotropic micro‐particle many‐body system can be investigated. Experimental investigations of electrorheological plasmas are performed under microgravity conditions in parabolic flights. For analysing the image data, a supervised machine learning code was developed and a continuous cross over was found. A molecular dynamics simulation showed qualitatively similar results but also some deviations from the experimental results. [ABSTRACT FROM AUTHOR]

While computing has been (re)introduced into the basic education curricula in various countries, its actual implementation appears to be inconsistent. There are schools in which computing education is commonplace, while the implementation seems to be lagging behind in others. There is emerging evidence that some teachers do not consider computing education relevant, meaningful and important and, thus, intentionally neglect its provision. This is problematic as understanding the principles of code and computing is crucial for agentic citizenship in the post‐digital era. This paper argues that one main reason for these teachers' reluctance is the economy‐driven discursive framing of computing education, which is in contrast with the socialization‐oriented manner in which teachers approach their work. To contribute to resolving this issue, the present paper introduces a transversal approach to computing education. It conceptualizes code as a sociomaterial text with social and societal histories and consequences. Theoretically and conceptually, the approach draws on the pedagogy of multiliteracies. The leading idea is that digital technologies are examined with students from functional and critical dimensions and through micro and macro perspectives. The use of wearable sports technologies, such as activity wristbands, are used as practical examples to put the theoretical ideas into context. Practitioner notesWhat is already known about this topic Computing has been (re)introduced in the curricula of basic education in various countries.Some teachers are found to be reluctant to teach computing in basic education.What this paper adds This paper introduces a transversal multiliteracies‐based approach for computing education.Implications for practice and policy Computing should be included in curricula and classrooms in a holistic manner that includes both functional and critical approaches to computing. [ABSTRACT FROM AUTHOR]

This study reduces the difficulties experienced by novice computer programmers when establishing a procedure for problem‐solving in a computing environment. It also provides guidelines to detail the work process. For this purpose, we developed an "unfolding model of procedures" consisting of the recognition of input and output, decomposition of operation, and expansion of branching based on its condition in work. To identify the effect of applying the unfolding model of procedures on the programming learning curve, we conducted an experiment with 110 college freshers randomly assigned to an experimental group and a control group. The unfolding model was applied to the experimental group; however, both groups received introductory programming classes for 5 weeks. The experimental group showed a trend of significant improvement in deriving the input and output processes and expanded their judgment on conditions. The evaluation of flowcharts for the same task conveyed that the experimental group was more competent than the control group, and further analysis by sex revealed no difference between the sexes in the achievement of learning within either group. The experimental group showed an advantage over the control group irrespective of sex. In conclusion, we expect that the unfolding model of procedures developed in this study could be used as a tool for novice programmers to derive work procedures completely and systematically when interpreting human work in a computing environment at the early stages of learning to program. [ABSTRACT FROM AUTHOR]

Although the importance of computational thinking (CT) has been widely recognized, we still poorly understand how to effectively teach this skill. The aim of this study is to investigate whether peer code review (PCR) could affect CT, learning engagement, and learning satisfaction in a blended learning environment. To achieve this, we carried out a randomized control trial study involving 84 first‐year undergraduates (second semester) from different majors. There was no difference in CT before this experiment and after the experiment, general CT, specific CT, learning engagement, and satisfaction were measured. An independent sample t test showed that students learning with code peer review showed better performance on computational concepts, computational perspectives, learning engagement, and learning satisfaction (p <.001). Bayesian t test further showed substantial or strong evidence for the improvement of computational concepts, learning engagement, learning satisfaction, and computational perspectives. Moreover, the absolute values of Cohen's effect sizes of computational concept, computational perspectives, emotional engagement, behavior engagement, learning engagement, learning satisfaction were all above 0.4, which is viewed as educationally meaningful. Although there was no difference in the general CT, we still saw evidence of improvement over the course of the experiment. Overall, our findings suggest instructors to use PCR reasonably to further improve CT, learning engagement, and satisfaction in programming courses. [ABSTRACT FROM AUTHOR]

Learning to code helps children to develop an understanding of how things work. It also contributes to the development of problem‐solving skills. In this paper, we present a methodological framework with four cycles that alternate between cooperative and individual learning approaches in addition to game‐based and project‐based learning. The presented framework is suitable for deployment in primary schools, regardless of the digital competencies and programming skills of the teachers. The framework was evaluated focusing on the approach (collaborative or competitive) that is more suitable for the first cycle in the multi‐cycle scenario for learning to code from a demographic perspective (gender and urbanization), the relevancy of Bloom's taxonomy, and the achieved learning experience. Initial results confirm that this approach encourages the students to use the proposed tools for finding the solution, raising their interest and motivation for learning to code. The teachers who participated in the pilot study were satisfied with the proposed approach, both in terms of achieved learning outcomes and the level of training needed to obtain necessary teaching competencies. [ABSTRACT FROM AUTHOR]

Even though it is well known that for most relevant computational problems, different algorithms may perform better on different classes of problem instances, most researchers still focus on determining a single best algorithmic configuration based on aggregate results such as the average. In this paper, we propose integer programming‐based approaches to build decision trees for the algorithm selection problem. These techniques allow the automation of three crucial decisions: (a) discerning the most important problem features to determine problem classes, (b) grouping the problems into classes, and (c) selecting the best algorithm configuration for each class. To evaluate this new approach, extensive computational experiments were executed using the linear programming algorithms implemented in the COIN‐OR branch‐and‐cut solver across a comprehensive set of instances, including all MIPLIB benchmark instances. The results exceeded our expectations. While selecting the single best parameter setting across all instances decreased the total running time by 22%, our approach decreased the total running time by 40% on average across 10‐fold cross‐validation experiments. These results indicate that our method generalizes quite well and does not overfit. [ABSTRACT FROM AUTHOR]

Humans are capable of acquiring multiple types of information presented in the same information stream. It has been suggested that at least two parallel learning processes are important during learning of sequential patterns—statistical learning and rule‐based learning. Yet, the neurophysiological underpinnings of these parallel learning processes are not fully understood. To differentiate between the simultaneous mechanisms at the single trial level, we apply a temporal EEG signal decomposition approach together with sLORETA source localization method to delineate whether distinct statistical and rule‐based learning codes can be distinguished in EEG data and can be related to distinct functional neuroanatomical structures. We demonstrate that concomitant but distinct aspects of information coded in the N2 time window play a role in these mechanisms: mismatch detection and response control underlie statistical learning and rule‐based learning, respectively, albeit with different levels of time‐sensitivity. Moreover, the effects of the two learning mechanisms in the different temporally decomposed clusters of neural activity also differed from each other in neural sources. Importantly, the right inferior frontal cortex (BA44) was specifically implicated in visuomotor statistical learning, confirming its role in the acquisition of transitional probabilities. In contrast, visuomotor rule‐based learning was associated with the prefrontal gyrus (BA6). The results show how simultaneous learning mechanisms operate at the neurophysiological level and are orchestrated by distinct prefrontal cortical areas. The current findings deepen our understanding on the mechanisms of how humans are capable of learning multiple types of information from the same stimulus stream in a parallel fashion. [ABSTRACT FROM AUTHOR]

Introduction and Aims: In October 2020, New Zealanders will vote on whether cannabis should be legalised for recreational use. With this in mind, the aim of the present study is to gauge the views and opinions of the New Zealand population on cannabis via tweets. To achieve this, we conducted a sentiment analysis of all historic cannabis‐related tweets and referendum‐specific tweets written in New Zealand. Design and Methods: We used a Twitter‐sponsored commercial platform to access all historic cannabis‐related tweets written in New Zealand and used search terms to remove non‐cannabis‐related terms. Next, we used the platform's machine learning function to code the sentiment of tweets (i.e. positive/pro‐cannabis, negative/anti‐cannabis or neutral). Results: Between July 2009 and August 2020, 304 760 cannabis‐related tweets were written in New Zealand. Overall, the tweets were predominantly positive (62.0%) and there was a higher proportion of positive tweets written in 2020 (65.3%) compared to negative or neutral tweets. Similarly, for referendum‐specific tweets, the 2020 data reveal a generally positive view of cannabis (53.5%). Discussion and Conclusions: Both cannabis‐related, and referendum‐specific tweets, suggest that Twitter users in New Zealand have a generally positive view of cannabis. Given the nature of Twitter, the current method will allow us to study whether views toward cannabis change as the referendum nears and capture any late swings in pro‐ or anti‐cannabis sentiment (abcd‐lab.shinyapps.io/cannabis_sentiment/). [ABSTRACT FROM AUTHOR]

Software project defect prediction can help developers allocate debugging resources. Existing software defect prediction models are usually based on machine learning methods, especially deep learning. Deep learning‐based methods tend to build end‐to‐end models that directly use source code‐based abstract syntax trees (ASTs) as input. They do not pay enough attention to the front‐end data representation. In this paper, we propose a new framework to represent source code called multiperspective tree embedding (MPT‐embedding), which is an unsupervised representation learning method. MPT‐embedding parses the nodes of ASTs from multiple perspectives and encodes the structural information of a tree into a vector sequence. Experiments on both cross‐project defect prediction (CPDP) and within‐project defect prediction (WPDP) show that, on average, MPT‐embedding provides improvements over the state‐of‐the‐art method. [ABSTRACT FROM AUTHOR]

An increasing number of countries have recently included programming education in their curricula. Similarly, utilizing programming concepts in gameplay has become popular in the videogame industry. Although many games have been developed for learning to program, their variety and their correspondence to national curricula remain an uncharted territory. Consequently, this paper has three objectives. Firstly, an investigation on the guidelines on programming education in K‐12 in seven countries was performed by collecting curricula and other relevant data official from governmental and non‐profit educational websites. Secondly, a review of existing acquirable games that utilize programming topics in their gameplay was conducted by searching popular game stores. Lastly, we compared the curricula and made suggestions as to which age group the identified games would be suitable. The results of this study can be useful to educators and curriculum designers who wish to gamify programming education. [ABSTRACT FROM AUTHOR]

Source code plagiarism detection using Running‐Karp‐Rabin Greedy‐String‐Tiling (RKRGST) is a common practice in academic environment. However, such approach is time‐inefficient (due to RKRGST's cubic time complexity) and insensitive (toward token subsequence rearrangement). This paper proposes ES‐Plag, a plagiarism detection tool featured with cosine‐based filtering and penalty mechanism to handle aforementioned issues. Cosine‐based filtering mitigates time‐inefficiency by excluding non‐potential pairs from RKRGST comparison; while penalty mechanism mitigates insensitivity by reducing the number of matched tokens with the number of matched subsequences prior similarity normalization. In addition to issue‐solving features, ES‐Plag is also featured with project‐based input, colorized adjacency similarity matrix, matched token highlighting, and various similarity algorithms (e.g., Cosine Similarity and Local Alignment). Three findings can be deducted from our evaluation. First, cosine‐based filtering boosts up time efficiency with a trade‐off in effectiveness. Second, penalty mechanism enhances sensitivity even though its improvement in terms of effectiveness is quite limited. Third, ES‐Plag's features are beneficial for examiners. [ABSTRACT FROM AUTHOR]

This article presents a continuous assessment methodology for a computer programming course supported by an automatic assessment tool, applied to the practical programming exercises performed by the students. The interaction between the students and the assessment tool was studied through quantitative analyses. In particular, the solutions proposed by the students (computer programs) were analyzed using the verdicts given by the automatic assessment tool: correct solutions or incorrect solutions. In the case of incorrect solutions, the types of programming errors were studied. Additionally, interaction was also studied by analyzing the students' success rate. This rate is the percentage of correct solutions among the total number of attempts (correct and incorrect). Moreover, the relationship between success rate and academic performance was analyzed. Furthermore, this research examines the students' perceptions toward the assessment tool through interviews. The results of this study help understanding the benefits and perceptions of the students with respect to the use of an automatic assessment tool in a computer programming course. [ABSTRACT FROM AUTHOR]

Abstract: Social networking sites (SNSs) are a popular Internet‐based means for users to communicate and interact with each other. Although they have caught the attention of many researchers and are already being used as educational tools, very few studies have investigated the effects of using an SNS in engineering education. This study, therefore, aims to analyze the effects of using the Edmodo platform as a teaching and learning support tool on students’ academic and practical performance in the Introduction to Information Technology and Algorithms course, as well as in the Computer Programming course they took in the following semester. It also considers the students’ opinions about the Edmodo system. For this study, a total of 62 students studying in the Electrical and Electronics Engineering Department during the 2016–2017 fall semester were divided into two equally sized groups. The control group underwent a traditional face‐to‐face education, whereas the experimental group augmented this using the Edmodo system. A mixed‐methods approach with a post‐test‐only control group design was used: quantitative data were obtained from student tests, together with qualitative data from follow‐up interviews. The students’ grades were analyzed using Student's t‐test and correlation analysis, showing that the experimental group performed better in their academic and laboratory assessments and that there was a moderately positive relationship between the post‐test results and performance in the subsequent Computer Programming course. [ABSTRACT FROM AUTHOR]

Abstract: The programming teaching faces serious challenges all over the world. Many students end up abandoning the Computer Science courses in their freshman years because they are not able to understand the concepts taught in introductory courses. On the other hand, the development of games is an opportunity for the students to perform an actual activity of the career and hence it may serve as an incentive to their motivation, commitment, and proactivity. This article describes a teaching approach based on the Problem Based Learning (PBL) methodology, which is associated to the development of educational games to drive the programming teaching in an introductory and practical course of programming teaching. The article presents details on the approach and how it was conducted in a case study with 30 students. The article also comprehends the approach's assessment, as well as the results got through the evaluation survey with students and teachers. [ABSTRACT FROM AUTHOR]

Educational computer programming in geometrical modeling is an interactive dynamic educational method which helps to visualize; interactively and graphically; 3D geometrical models under different design conditions. Students using geometrical modeling in hydraulics can get large amount of geometrical data describing the x, y, and z coordinates of any circular cross-section of the designed 3D canal transition, a step which is very important in the modeling of field and laboratory 3D canal sections. They can visually and interactively perform mathematical analysis of very complicated models very easily. Educational programming of geometrical modeling is very similar to experimental modeling in which the students gets feedback from the computer program and adjust--if necessary--the model specifications and properties to get the best engineering design. In this paper, the LabVIEW Platform and the 3D Bezier surface design method will be used by the students, as an educational interactive computer programming, for the design of a 3D open circular canal crosssection having a hydraulic constriction and, a bottom slope of certain dimensions. [ABSTRACT FROM AUTHOR]

Abstract: Logic Programming (LP) follows the declarative programming paradigm, which novice students often find hard to grasp. The limited availability of visual teaching aids for LP can lead to low motivation for learning. In this paper, we present a platform for teaching and learning Prolog in Virtual Worlds, which enables the visual interpretation and verification of program results in a straightforward fashion and requires students to adopt a collaborative problem‐solving approach. The results of the pilot application and student‐centered evaluation of the platform are encouraging regarding group learning performance and user experience. The paper contributes to current practice of teaching and learning LP by proposing a metaphor and a system that can empower the educational process with toy world examples visualized in a shared 3D environment. [ABSTRACT FROM AUTHOR]

Abstract: Crowdcoding is a programming model that outsources a software project implementation to the crowd. As educators, we think that crowdcoding could be leveraged as part of the learning path of engineering students from a computer programming introductory course to solve local community problems. The benefits are twofold: on the one hand the students practice the concepts learned in class and, on the other hand, they participate in real‐life problems. Nevertheless, several challenges arise when developing a crowdcoding platform, the first one being how to check the correctness of student's code without giving an extra burden to the professors in the course. To overcome this issue, we propose a novel system that does not resort to expert review; neither requires knowing the right answers beforehand. The proposed scheme automatically clusters the student's codes based solely on the output they produce. Our initial results show that the largest cluster contains the same codes selected as correct by the automated and human testing, as long as some conditions apply. [ABSTRACT FROM AUTHOR]

This paper presents the results of a systematic review of the literature, including a meta-analysis, about the effectiveness of the use of Alice software in programming learning when compared to the use of a conventional programming language. Our research included studies published between the years 2000 and 2014 in the main databases. We gathered 232 papers. Taking into account the selection criteria to make the meta-analysis, we retained six papers with a quasi-experimental design, with 464 participants in total. To combine the results we used the random effect model. It resulted in an effect size of 0.54 (Cohen's d) with a confidence interval between 0.34 and 0.74. We concluded that until now there have been few experimental results on the effectiveness of Alice programming language to introduce students in learning how to program. The results we found were the expression of different experimental treatments and distinguished teaching methods which made the comparison of the results obtained more subtle. However, the existing experimental results that were submitted to the meta-analysis allowed us to assume with a certain margin of safety that a teaching strategy that uses Alice should obtain more effective results than the use of a conventional programming language. [ABSTRACT FROM AUTHOR]

ABSTRACT Algorithms are important procedures that are found in every aspect in our life. Large numbers of these algorithms cannot be accurately understood unless they are presented differently, especially if they are taught to students. A complete visualization system is created to teach dynamic programming algorithms. Through visualization, a student is able to trace the algorithm step-by-step, similar to the debugger, but in a simplified way. In addition, the system can handle errors during the execution of the algorithm. Moreover, the system guides and helps the students during the tracing and stores all of the information about the errors during the tracing. The students obtain the feedback of the tracing from a report generated by the system. The teacher benefits from the generated report through the enhancement of the teaching methods that focus on the misunderstood steps. Additionally, the system allows the user to see how a certain algorithm can be monitored by displaying the source code. The developed system is evaluated using different methods to prove the effectiveness of the system in teaching and learning.© 2017 Wiley Periodicals, Inc. Comput Appl Eng Educ 25:179-187, 2017; View this article online at ; DOI . [ABSTRACT FROM AUTHOR]

International changes in policy and curricula (notably recent developments in England) have led to a focus on the role of computing education in the early years. As interest in the potential of computing education has increased, there has been a proliferation of programming tools designed for young children. While these changes are broadly to be welcomed, the pace of change has arguably led to more attention to the tools than to key questions about pedagogy. This paper proposes three areas of research (Logo; computational thinking; and teaching STEM subjects of science, technology, engineering and mathematics) that may inform computing education for young children and suggests that a greater focus on thinking skills and connections to manifestations of computers in the real world is needed. Above all, the paper calls for an informed debate about the trend towards introducing computing education to children in the early years. [ABSTRACT FROM AUTHOR]

Recent government moves in many countries have seen coding included in school curricula, or promoted as part of computing, mathematics or science programmes. While these moves have generally been associated with a need to engage more young people in technology study, research has hinted at possible benefits from learning to program including fostering general thinking skills. However, little research has been carried out exploring these ideas. This study analysed data collected while 5- and 6-year-old students in a New Zealand primary school were using Scratch Jnr. to learn about basic shapes, as part of a numeracy topic. Analysis combined Brennan and Resnick's (2012) computational thinking skills framework and Krathwohl's (2002) revision of Bloom's Taxonomy to evaluate any role general thinking skills played in these students' coding work. Results suggest including basic coding in primary curricula provides teachers with an effective means of exercising their students' general and higher order thinking skills. They build on Brennan and Resnick's (2012) framework by including conceptualization as an important element in students' computational work and highlight the role of predictive thinking in debugging code. Findings support historical arguments that more needs to be done to investigate students' cognitive processes when undertaking computational work. Lay description What is already known about the topic: Computational learning is an emerging area of school curricula;, Limited research exists exploring thinking processes within computational learning;, Early studies challenge more recent claims of thinking skill transfer from computational work., What this paper adds: Computational work supports a range of general and higher order thinking skills;, Task design and teacher skills are critical to achieving higher order thinking outcomes from computational work;, Computational work in teams can support collaborative, cooperative and self-management key competencies., Implications for practice and/or policy: Findings broaden the base of empirical support for including computational work in school curricula;, Coding provides an engaging means of exercising complex thinking skills and key competencies in students;, The data methods used provide teachers with visible evidence of students' thinking processes during computational work. [ABSTRACT FROM AUTHOR]

The purpose of this study is to evaluate the effect of using Facebook on enriching the learning experiences of students in programming learning. In another word, the aim of the study is to detect the reflections from the environment of programming learning where Facebook is used. The study was carried out within the scope of the Introduction to Programming and Algorithm course taught in the software engineering department of the technology faculty of a university in Turkey. Initially, a Facebook group was created to enrich classroom teaching. Sixty-three freshman engineering students taking the course and the faculty members of the department joined the group. The study was conducted for one semester in the fall semester of the 2014-2015 academic year. The Educational Usage of Facebook Scale (EUFS) and semi-structured interview formwere used as data collection tools in the study. Quantitative data were descriptively analyzed while qualitative data were subjected to content analysis. The results obtained in this study show that Facebook has positive effects such as providing a rich content in programming learning, increasing extracurricular communication, offering a cooperative and collaborative learning environment, increasing motivation, and bringing different perspectives as well as negative effects such as causing distraction, decreasing face-to-face communication, and bringing about problems about expressing oneself. As a result, it is possible to assert that Facebook can be efficiently used to support classroom teaching strategies and enrich learning experiences in programming learning. [ABSTRACT FROM AUTHOR]

ABSTRACT This paper presents an online C compiler designed so that students can program their practical assignments in Programming courses. What is really innovative is the self-assessment of the exercises based on black-box tests and train students' skill to test software. Moreover, this tool lets instructors, not only proposing and classifying practical exercises, but also evaluating automatically the efforts dedicated and the results obtained by the students. The system has been applied to the 1st-year students at the Industrial Engineering specialization at the Universidad Politecnica de Madrid. Results show that the students obtained better academic performance, reducing the failure rate in the practical exam considerably with respect to previous years, in addition that an anonymous survey proved that students are satisfied with the system because they get instant feedback about their programs. © 2015 Wiley Periodicals, Inc. Comput Appl Eng Educ 23:887-896, 2015; View this article online at ; DOI . [ABSTRACT FROM AUTHOR]

This article presents a proposal for the detection of programming source code similitude in academic environments. The objective of this proposal is to provide support to professors in detecting plagiarism in student homework assignments in introductory computer programming courses. The developed tool, CODESIGHT, is based on a modification of the Greedy String Tiling algorithm. The tool was tested in one theoretical and three real scenarios, obtaining similitude detections for assignments ranging from those that contained code without modifications to assignments containing insertions of procedural instructions inside the main code. The results verified the efficiency of the tool at the first five levels of the plagiarism spectrum for programming code, in addition to supporting suspicions of plagiarism in real scenarios. © 2013 Wiley Periodicals, Inc. Comput Appl Eng Educ 23:13-22, 2015; View this article online at ; DOI [ABSTRACT FROM AUTHOR]

ABSTRACT This practical report analyzes a programming class using a micro robot (MR), the smallest soccer robot in the RoboCup world competition. This class examined the effect of using the MR as the actual equipment employed in programming. Questionnaire results on this class revealed that these teaching materials evoked a heightened programming interest among students. Moreover, the problems related to programming instruction using an MR were better understood; therefore, a strategy for improving the related problems is discussed here. © 2013 Wiley Periodicals, Inc. Comput Appl Eng Educ 23:109-116, 2015; View this article online at ; DOI [ABSTRACT FROM AUTHOR]

This study examined whether developing earlier forms of knowledge in specific learning environments prepares students better for future learning when they are placed in an unfamiliar learning environment. Forty-one students in the fifth and sixth grades learned to program robot movements using abstract concepts of speed, distance and direction. Students in high-transparency environments learned visual programming to control robots (eg, organizing visual icons), and students in low-transparency environments learned syntactic programming to control robots (eg, text-based coding). Both groups received feedback and models of solutions during the learning phase. The assessment midway showed students in both conditions learned equally well when solving problems using familiar materials. However, a difference emerged when students were asked to solve new problems, using unfamiliar materials. The low-transparency group was more successful in adapting and repurposing their knowledge to solve novel problems that required the use of unfamiliar high-transparency materials. Students in the high-transparency group were less successful in adapting their knowledge when solving new problems using unfamiliar low-transparency materials. Both groups then proceeded to learn in the opposing transparency environments. The posttest revealed the benefits of initial learning in low-transparency environments as students performed better on repeated and new inferential problems across virtual and physical platforms. [ABSTRACT FROM AUTHOR]

The purpose of this study was to investigate the influence of goal specificity and scaffolding on the programming performance and self-regulation of elementary students engaged in learning game design. This study recruited 232 students for the experimental activities. Two levels of goal specificity were employed: specific and nonspecific. Structuring and problematizing scaffolds were also used to support students in the process of learning. Our results reveal the following: (1) nonspecific goals and structuring scaffolds led to superior comprehension of programming, (2) nonspecific goals with problematizing scaffolds led to better problem-solving performance, (3) problematizing scaffolds were more effective than structuring scaffolds in the development of self-regulation behavior, and (4) nonspecific goals were more effective than specific goals in the promotion of resource management associated with self-regulation. [ABSTRACT FROM AUTHOR]

This column examines a case study focusing on web design as an example of interest-driven learning and the acquisition of (digital media) literacies. A summer workshop was offered at a working-class public library, led by a self-taught seventeen year old girl. Nine students (ages 8-16) learned basic HTML and CSS and designed their own websites in a three week course. The workshop was grounded in an interest-driven approach to learning that focused on participants' identities and passions. Code was presented as a way they could express their interests and share them with a wider audience. In addition to learning digital media literacies, participants also practiced problem solving skills, creative thinking, and collaboration. The structure was informal and encouraged experimentation, failure, play, and mentorship. After the workshop, seven out of the nine participants expressed an interest in further pursuing code and a newfound interest in computers and technology. [ABSTRACT FROM AUTHOR]

In this study, the score equivalence of computer-based assessment (CBA) and paper-and-pencil testing (PPT) in an undergraduate course of introductory computer programing at the National Technical University of Athens' School of Chemical Engineering, was examined. The research was carried out with 211 students participating, and was realized by a specially constructed CBA platform as well as the classical PPT. Both types of assessment included the same questions of multiple-choice type, completing and correcting code in Fortran, as well as writing Fortran code from scratch. The results showed that, in general, students performed better and received higher scores under CBA compared to PPT. In particular, the very good students benefited from the advantages that the computer compiler offers in debugging code and completing code exercises and, thus, received relatively higher scores in these parts of the test. Moreover, switching answers and devoting more time-on-task was found to be beneficial for the students' performance. Overall, the findings of this study revealed that the actual programing abilities of the students are better evaluated using such CBA tests. © 2011 Wiley Periodicals, Inc. Comput Appl Eng Educ 21: 671-683, 2013 [ABSTRACT FROM AUTHOR]

Program tracing is an important skill for programming learning, but novice programmers' tracing abilities are poor. This study designs a system to enable students to exercise program tracing and make students' program tracing processes visible to help teachers indicate students' errors in tracing execution flow, variable values, or I/O display. © 2010 Wiley Periodicals, Inc. Comput Appl Eng Educ 21: 432-438, 2013 [ABSTRACT FROM AUTHOR]

This paper compares two different approaches for teaching complex programming concepts in lab sessions. The first approach is based on the widely held assumption that the programming knowledge and skills which the student obtains are predominantly acquired from classroom activities undertaken at the instigation of the teacher. This automatically implies that the student must attend a minimum number of lab sessions. The second approach is to motivate student learning via a full-term activity based on the implementation of computer games from scratch. © 2010 Wiley Periodicals, Inc. Comput Appl Eng Educ 21: 237-244, 2013 [ABSTRACT FROM AUTHOR]

At the nexus of education and technology, blended learning is growing rapidly. Integrating face-to-face and online learning, blending can enhance learning and optimize seat time. This paper describes the use of the blended e-learning model in a course 'Data Structures and Algorithms' given at the Afyon Kocatepe University, Turkey. This model is realized as a combination of a face-to-face environment and online learning, using the University's Learning Management System (LMS) named @KU-UZEM. The LMS consists of many applications in accordance with SCORM standards, such as, student records, user roles, courses, exams security applications, student affairs, counseling services, internal communication, director processes and evaluation. It provides whole software infrastructure by a virtual academic institution and is currently used in Afyon Kocatepe University. This paper describes both the technology for course design and programme redesign adopted a blended learning approach with both face-to-face and online learning aimed at enhancing the students' control over their own learning. According to the obtained results with the performed experimental evaluation, the realized blended learning model provided more effective and efficient educational experience rather than traditional, face-to-face learning. A survey conducted at the end of the course also showed that students were satisfied with the pedagogical approach, and their academic achievements were also better than expected. Particularly important is that the dropout rate was greatly diminished, which could be related to students' satisfaction with the support they received from the teacher and the system. © 2010 Wiley Periodicals, Inc. Comput Appl Eng Educ 21: 328-342, 2013 [ABSTRACT FROM AUTHOR]

SUMMARY Automated software verification is an active field of research, which has made enormous progress both in theoretical and practical aspects. Even if not ready for large-scale industrial adoption, the technology behind automated program verifiers is now mature enough to gracefully handle the kind of programs that arise in introductory programming courses. This opens exciting new opportunities in teaching the basics of reasoning about program correctness to novice students. However, for these tools to be effective, command-line-style user-interfaces need to be replaced. In this paper, we report on our experience using the verifying compiler for PEST in an introductory programming course as well as in a more advanced course on program analysis. PEST is an extremely basic programming language, but with expressive annotations capabilities and semantics amenable to verification. In particular, we comment on the crucial role played by the integration of this verifying compiler with the Eclipse integrated development environment. Copyright © 2012 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]