Structural development defects essentially refer to code structure that violates object-oriented design principles. They make program maintenance challenging and deteriorate software quality over time. Various detecti...Structural development defects essentially refer to code structure that violates object-oriented design principles. They make program maintenance challenging and deteriorate software quality over time. Various detection approaches, ranging from traditional heuristic algorithms to machine learning methods, are used to identify these defects. Ensemble learning methods have strengthened the detection of these defects. However, existing approaches do not simultaneously exploit the capabilities of extracting relevant features from pre-trained models and the performance of neural networks for the classification task. Therefore, our goal has been to design a model that combines a pre-trained model to extract relevant features from code excerpts through transfer learning and a bagging method with a base estimator, a dense neural network, for defect classification. To achieve this, we composed multiple samples of the same size with replacements from the imbalanced dataset MLCQ1. For all the samples, we used the CodeT5-small variant to extract features and trained a bagging method with the neural network Roberta Classification Head to classify defects based on these features. We then compared this model to RandomForest, one of the ensemble methods that yields good results. Our experiments showed that the number of base estimators to use for bagging depends on the defect to be detected. Next, we observed that it was not necessary to use a data balancing technique with our model when the imbalance rate was 23%. Finally, for blob detection, RandomForest had a median MCC value of 0.36 compared to 0.12 for our method. However, our method was predominant in Long Method detection with a median MCC value of 0.53 compared to 0.42 for RandomForest. These results suggest that the performance of ensemble methods in detecting structural development defects is dependent on specific defects.展开更多
Many design engineers in cross-domain industries have attended training classes of TRIZ to improve their innovative abilities in China. Most of them are successful, but others are not. So the latest target of the trai...Many design engineers in cross-domain industries have attended training classes of TRIZ to improve their innovative abilities in China. Most of them are successful, but others are not. So the latest target of the trainers is to improve the training process used now in industry in China and to make the engineers to understand the basic principles of TRIZ better. Based on the mass-engineer-oriented training model(MEOTM) and mechanical engineers’ design cases, a relationship between managing activities about the opportunities for innovation and the training process is set up. It is shown that the inventive problems come first from opportunity searching for engineers. A training and gate system for evaluation is developed to involve the managing activities of the companies in the training process. Then comparison between the general analogous process and the application of TRIZ is made, which shows the advantages and depth principles of TRIZ for the engineers to apply them confidently. Lastly a new process is formed in which opportunity searching, engineers training, inventive problems identifying and solving,and three redesign paths are connected seamlessly. The research proposes an opportunity-driven redesign path that cooperates the training and opportunity searching, which will be applied in future training classes to make more and more engineers to follow.展开更多
背景农村订单定向医学生(简称定向医学生)的免费培养工作是缓解我国农村地区基层医生短缺和提升基层医疗卫生队伍整体水平的重要举措,既往研究对我国各院校定向医学生教育研究的现状、方法及其效果评价的综合性分析不足。目的探讨2010—...背景农村订单定向医学生(简称定向医学生)的免费培养工作是缓解我国农村地区基层医生短缺和提升基层医疗卫生队伍整体水平的重要举措,既往研究对我国各院校定向医学生教育研究的现状、方法及其效果评价的综合性分析不足。目的探讨2010—2023年针对我国五年制本科定向医学生的医学教育研究的发展现状、研究质量和未来趋势,为后续工作提供参考。方法在中国知网、万方数据知识服务平台、维普网、PubScholar公益学术平台、PubMed、Web of Science、Cochrane Library共7个中英文数据库中检索2010—2023年收录的关于定向医学生培养的文献,由2名研究者根据文献纳入和排除标准,独立筛选文献、提取资料后,利用医学教育研究质量评价表(MERSQI)和纽卡斯尔·渥太华量表-教育版(NOS-E)对文献进行质量评价和综合分析,采用描述分析法对结果进行汇总、分析。结果共纳入文献37篇,其中中文36篇,英文1篇。研究设计中,对照前测-后测的研究设计占比最大(46%,17/37),单组后测和随机对照后测各占比22%(8/37,8/37),单组前测-后测占比8%(3/37)。97%的研究是围绕本科院校教育阶段开展的(36/37)。研究关注的内容依次是课程调整类(89%,33/37)、教学方法调整类(81%,30/37)、培养模式构建类(8%,3/37),其中,3项(8%,3/37)培养模式构建类研究和7项(19%,7/37)课程调整类研究开设专门针对农村、基层或全科的课程。结局评估主要针对学生对教学的评价(70%,26/37)和教学后的知识和技能提升(86%,32/37),较少涵盖实际行为改变(3%,1/37)及患者和医疗机构的获益(3%,1/37)。总体研究质量不高,MERSQI总分(10.4±2.4)分,最高分14.0分,其中样本机构数、评价工具效度、结局指标是影响低分的主要条目;NOS-E总分(2.5±1.5)分,最高分5.0分,其中对照组可比性、盲法是影响低分的主要条目。结论定向医学生的教育培训研究关注度增加,但存在研究质量普遍不高、跨机构跨地区研究不足、研究内容和评价指标缺乏订单定向特色、对毕业后教育和继续教育关注不足等问题。建议未来应加强多机构合作和国际交流,提高研究设计和方法的质量加强多机构合作研究及国际交流,提高研究设计与方法质量,并建立具有订单定向特色的统一评价体系;同时,加强农村及全科相关的订单定向特色课程设计,关注包含毕业后教育和继续教育阶段的连续性一体化教学设计。展开更多
文摘Structural development defects essentially refer to code structure that violates object-oriented design principles. They make program maintenance challenging and deteriorate software quality over time. Various detection approaches, ranging from traditional heuristic algorithms to machine learning methods, are used to identify these defects. Ensemble learning methods have strengthened the detection of these defects. However, existing approaches do not simultaneously exploit the capabilities of extracting relevant features from pre-trained models and the performance of neural networks for the classification task. Therefore, our goal has been to design a model that combines a pre-trained model to extract relevant features from code excerpts through transfer learning and a bagging method with a base estimator, a dense neural network, for defect classification. To achieve this, we composed multiple samples of the same size with replacements from the imbalanced dataset MLCQ1. For all the samples, we used the CodeT5-small variant to extract features and trained a bagging method with the neural network Roberta Classification Head to classify defects based on these features. We then compared this model to RandomForest, one of the ensemble methods that yields good results. Our experiments showed that the number of base estimators to use for bagging depends on the defect to be detected. Next, we observed that it was not necessary to use a data balancing technique with our model when the imbalance rate was 23%. Finally, for blob detection, RandomForest had a median MCC value of 0.36 compared to 0.12 for our method. However, our method was predominant in Long Method detection with a median MCC value of 0.53 compared to 0.42 for RandomForest. These results suggest that the performance of ensemble methods in detecting structural development defects is dependent on specific defects.
基金Supported by National Natural Science Foundation of China(Grant Nos.51675159,51305123)
文摘Many design engineers in cross-domain industries have attended training classes of TRIZ to improve their innovative abilities in China. Most of them are successful, but others are not. So the latest target of the trainers is to improve the training process used now in industry in China and to make the engineers to understand the basic principles of TRIZ better. Based on the mass-engineer-oriented training model(MEOTM) and mechanical engineers’ design cases, a relationship between managing activities about the opportunities for innovation and the training process is set up. It is shown that the inventive problems come first from opportunity searching for engineers. A training and gate system for evaluation is developed to involve the managing activities of the companies in the training process. Then comparison between the general analogous process and the application of TRIZ is made, which shows the advantages and depth principles of TRIZ for the engineers to apply them confidently. Lastly a new process is formed in which opportunity searching, engineers training, inventive problems identifying and solving,and three redesign paths are connected seamlessly. The research proposes an opportunity-driven redesign path that cooperates the training and opportunity searching, which will be applied in future training classes to make more and more engineers to follow.
文摘背景农村订单定向医学生(简称定向医学生)的免费培养工作是缓解我国农村地区基层医生短缺和提升基层医疗卫生队伍整体水平的重要举措,既往研究对我国各院校定向医学生教育研究的现状、方法及其效果评价的综合性分析不足。目的探讨2010—2023年针对我国五年制本科定向医学生的医学教育研究的发展现状、研究质量和未来趋势,为后续工作提供参考。方法在中国知网、万方数据知识服务平台、维普网、PubScholar公益学术平台、PubMed、Web of Science、Cochrane Library共7个中英文数据库中检索2010—2023年收录的关于定向医学生培养的文献,由2名研究者根据文献纳入和排除标准,独立筛选文献、提取资料后,利用医学教育研究质量评价表(MERSQI)和纽卡斯尔·渥太华量表-教育版(NOS-E)对文献进行质量评价和综合分析,采用描述分析法对结果进行汇总、分析。结果共纳入文献37篇,其中中文36篇,英文1篇。研究设计中,对照前测-后测的研究设计占比最大(46%,17/37),单组后测和随机对照后测各占比22%(8/37,8/37),单组前测-后测占比8%(3/37)。97%的研究是围绕本科院校教育阶段开展的(36/37)。研究关注的内容依次是课程调整类(89%,33/37)、教学方法调整类(81%,30/37)、培养模式构建类(8%,3/37),其中,3项(8%,3/37)培养模式构建类研究和7项(19%,7/37)课程调整类研究开设专门针对农村、基层或全科的课程。结局评估主要针对学生对教学的评价(70%,26/37)和教学后的知识和技能提升(86%,32/37),较少涵盖实际行为改变(3%,1/37)及患者和医疗机构的获益(3%,1/37)。总体研究质量不高,MERSQI总分(10.4±2.4)分,最高分14.0分,其中样本机构数、评价工具效度、结局指标是影响低分的主要条目;NOS-E总分(2.5±1.5)分,最高分5.0分,其中对照组可比性、盲法是影响低分的主要条目。结论定向医学生的教育培训研究关注度增加,但存在研究质量普遍不高、跨机构跨地区研究不足、研究内容和评价指标缺乏订单定向特色、对毕业后教育和继续教育关注不足等问题。建议未来应加强多机构合作和国际交流,提高研究设计和方法的质量加强多机构合作研究及国际交流,提高研究设计与方法质量,并建立具有订单定向特色的统一评价体系;同时,加强农村及全科相关的订单定向特色课程设计,关注包含毕业后教育和继续教育阶段的连续性一体化教学设计。
