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
Serious desertification caused by human activity and climate change,in addition to water loss and soil erosion related to arsenic sandstone in the Mu Us Sandy Land,lead to severe scarcity of soil and water resources,w...Serious desertification caused by human activity and climate change,in addition to water loss and soil erosion related to arsenic sandstone in the Mu Us Sandy Land,lead to severe scarcity of soil and water resources,which causes worse local agricultural conditions accordingly.Many physical properties of arsenic sandstone is complementary with that of sand,arsenic sandstone is therefore supposed to be blended to enhance water productivity and arability of sandy land.Container experiments are carried out to study the enhancement of water holding capacity of the mixture,the blending ratio of arsenic sandstone and sand,and the proper size of the arsenic sandstone particles,respectively.The results of the experiments show that particle size of 4 cm with a ratio of 1∶2 between arsenic sandstone and sand are the proper parameters on blending.Both water content and fertility increase after blending.Water use efficiency in the mixture is 2.7 times higher than that in sand by the water release curves from experiments.Therefore,a new sand control and development model,including arsenic sandstone blending with sand,efficient water irrigation management and reasonable farming system,is put forward to control and develop sandy land so that water-saving agriculture could be developed.Demonstration of potato planting about 153.1 ha in area in the Mu Us Sandy Land in China indicates that water consumption is 3018 m3/ha in the whole growth period.It means that about 61%of irrigation water can be saved compared with water use in coarse sand without treatment.Recycle economic mode and positive feedback of sand resource-crop planting-soil resource are constructed,which changes sand into arable soil and make it possible to develop water-saving agriculture on it.The proposed model will be helpful for soil-water resources utilization and management in the Mu Us Sandy Land.展开更多
The CTB Water Wall project is a maximal product life cycle utilization concept study by members of the space architecture design community.Its function is to demonstrate a human space activity Cargo Transport Bag(CTB)...The CTB Water Wall project is a maximal product life cycle utilization concept study by members of the space architecture design community.Its function is to demonstrate a human space activity Cargo Transport Bag(CTB)that becomes a primary water recycling membrane element after delivery of cargo,and then a permanent architectural building block for sustainable space habitation after its use in water treatment is complete.As such,it is intended as an experiment in radical life cycle product optimization in an extremely mass-constrained application environment(human space operations).It also introduces some fundamentally interesting concepts in architectural use of waste materials in extreme environments.Finally,it is in some ways a simple,tactile and visual demonstration of how far sustainable product design can be taken,if the motivation and technical justification are present.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Under the auspices of National Natural Science Foundation of China(No.51079120)Education Department Research Program of Shaanxi Province(No.12JK0481)Water Conservancy Science and Technology Plan of Shaanxi Province(No.2012-07)
文摘Serious desertification caused by human activity and climate change,in addition to water loss and soil erosion related to arsenic sandstone in the Mu Us Sandy Land,lead to severe scarcity of soil and water resources,which causes worse local agricultural conditions accordingly.Many physical properties of arsenic sandstone is complementary with that of sand,arsenic sandstone is therefore supposed to be blended to enhance water productivity and arability of sandy land.Container experiments are carried out to study the enhancement of water holding capacity of the mixture,the blending ratio of arsenic sandstone and sand,and the proper size of the arsenic sandstone particles,respectively.The results of the experiments show that particle size of 4 cm with a ratio of 1∶2 between arsenic sandstone and sand are the proper parameters on blending.Both water content and fertility increase after blending.Water use efficiency in the mixture is 2.7 times higher than that in sand by the water release curves from experiments.Therefore,a new sand control and development model,including arsenic sandstone blending with sand,efficient water irrigation management and reasonable farming system,is put forward to control and develop sandy land so that water-saving agriculture could be developed.Demonstration of potato planting about 153.1 ha in area in the Mu Us Sandy Land in China indicates that water consumption is 3018 m3/ha in the whole growth period.It means that about 61%of irrigation water can be saved compared with water use in coarse sand without treatment.Recycle economic mode and positive feedback of sand resource-crop planting-soil resource are constructed,which changes sand into arable soil and make it possible to develop water-saving agriculture on it.The proposed model will be helpful for soil-water resources utilization and management in the Mu Us Sandy Land.
文摘The CTB Water Wall project is a maximal product life cycle utilization concept study by members of the space architecture design community.Its function is to demonstrate a human space activity Cargo Transport Bag(CTB)that becomes a primary water recycling membrane element after delivery of cargo,and then a permanent architectural building block for sustainable space habitation after its use in water treatment is complete.As such,it is intended as an experiment in radical life cycle product optimization in an extremely mass-constrained application environment(human space operations).It also introduces some fundamentally interesting concepts in architectural use of waste materials in extreme environments.Finally,it is in some ways a simple,tactile and visual demonstration of how far sustainable product design can be taken,if the motivation and technical justification are present.
