An Approach to Detect Structural Development Defects in Object-Oriented Programs

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.

基于MobileNet-SSD的铁路信号灯检测识别

为解决铁路轨旁信号灯的定位与实时检测问题,在深入分析传统SSD算法与MobileNet模型的基础上,将MobileNet模型的最后平均池化层、全连接层转换为SSD算法的多尺度特征映射层,提出了一种基于MobileNet-SSD的铁路信号灯检测算法.实验结果表明:该算法克服了传统SSD算法对小目标识别不准确、检测实时性较差的问题,检测速度更快,准确率更高;在50 m监控范围内,算法的平均检测准确率达到85%以上,同时具有25帧/s的实时识别能力.

Preliminary Study of Chemical Elements Distribution in Petroleum Source Rocks Donga and Yogou Formations of the Termit Sedimentary Basin (Est-Niger)

XRF and EDX analyses were carried out on 18 batches of representative raw samples to determine the distribution of major chemical elements in the petroleum source rocks of Donga and Yogou formations of Termit sedimentary basin. The chemical composition of these formations is dominated by silicon (Si), aluminum (Al) and iron (Fe). This is consistent with the oxide composition, which is also dominated by silicon oxide (SiO2), aluminum oxide (Al2O3) and iron monoxide (FeO). No less important chemical elements are calcium (Ca), potassium (K), sulfur (S), titanium (Ti), magnesium (Mg), manganese (Mn) and barium (Ba), as well as some of their oxides. All these major chemical elements are carried by silicate detrital minerals associated with pyrite and goethite and/or clay minerals such as kaolinite and interstratified illite, smectite and chlorite. This trend is illustrated by the values of the Si/Al and SiO2/Al2O3 ratios.

基于T-S模糊故障树的制动系统可靠性分析

为解决动车组制动系统由于工作条件多样及内部结构复杂而呈现出的多故障状态以及故障机理模糊、故障关系不确定等问题,应用T-S模糊故障树的逻辑关系明确且可处理事件的不确定能力,结合基于信心指数修正的专家调查法,进行制动系统模型构建及重要度计算。得出了动车组制动系统薄弱环节,可为系统检修提供参考。

rt-PA静脉溶栓联合灯盏细辛注射液治疗急性缺血性脑卒中的效果分析

目的:探讨用重组组织型纤溶酶原激活剂(rt-PA)静脉溶栓疗法联合灯盏细辛注射液治疗急性缺血性脑卒中的临床效果。方法:选取2018年1月至2019年6月佛山市中医院卒中中心收治的急性缺血性脑卒中患者84例(行rt-PA静脉溶栓治疗),对其临床资料进行回顾性研究。按是否使用灯盏细辛注射液进行治疗将其分为观察组(46例)及对照组(38例),比较两组患者的临床疗效、NIHSS评分、Barthel评分和不良反应的发生情况。结果:观察组患者治疗的总有效率明显高于对照组患者,P<0.05。治疗后,两组患者的NIHSS评分均较治疗前显著降低,P<0.05;观察组患者的NIHSS评分低于对照组患者,P<0.05。治疗后,两组患者的Barthel评分均较治疗前显著提高,P<0.05;观察组患者的Barthel评分高于对照组患者,P<0.05。治疗后,两组患者均未出现血压降低、头痛、过敏、胸闷、心悸等不良反应,其血尿常规指标、肝肾功能指标及电解质指标均无异常。结论:用rt-PA静脉溶栓疗法联合灯盏细辛注射液治疗急性缺血性脑卒中的临床效果显著,可有效地减轻患者的临床症状,改善其神经功能,提高其日常生活活动能力和生活质量,且安全性较高。

含沙量监测的wavelet-Kalman多尺度融合研究

为解决黄河含沙量监测时传感器易受环境因素影响的问题,简述了音频共振法测量含沙量的原理,探讨了音频共振传感器的谐振频率和含沙量监测之间的关系,提出了贯序式wavelet-Kalman多尺度融合模型,对音频共振传感器的谐振频率进行小波分解,把含沙量数据组成贯序式数据块进行多尺度分析,提取含沙量信号序列中的突变值,建立了Kalman融合方程,将温度信息作为控制信号,消除了环境因素对含沙量监测的影响,并进行了含沙量测量的反演和误差分析。结果表明:贯序式wavelet-Kalman多尺度融合模型能够有效地消除环境影响,提高系统测量的精度和稳定性,平均绝对误差为3.95 kg/m^3,均方根误差为3.13 kg/m^3,比其他反演模型的误差小。

Handling Qualities of a New Last-Mile Vehicle

This paper assesses the vehicle dynamics of a new cargo bike concept developed for euro pallet sized cargo. The cargo bike developed is for last-mile delivery. Different aspects of manoeuvrability and stability are examined using a series of manoeuvres based on tests from the automotive industry combined with bicycle industry regulations. These manoeuvres objectively evaluate and determine the handling capabilities of the cargo bike concept. Those tests can be compared using the results of the benchmark vehicles. The results conclude the new cargo bike has proper vehicle dynamics above the majority of benchmark vehicles but there is still room for improvement.

Designing a High-Performance Deep Learning Theoretical Model for Biomedical Image Segmentation by Using Key Elements of the Latest U-Net-Based Architectures

Deep learning (DL) has experienced an exponential development in recent years, with major impact in many medical fields, especially in the field of medical image and, respectively, as a specific task, in the segmentation of the medical image. We aim to create a computer assisted diagnostic method, optimized by the use of deep learning (DL) and validated by a randomized controlled clinical trial, is a highly automated tool for diagnosing and staging precancerous and cervical cancer and thyroid cancers. We aim to design a high-performance deep learning model, combined from convolutional neural network (U-Net)-based architectures, for segmentation of the medical image that is independent of the type of organs/tissues, dimensions or type of image (2D/3D) and to validate the DL model in a randomized, controlled clinical trial. We used as a methodology primarily the analysis of U-Net-based architectures to identify the key elements that we considered important in the design and optimization of the combined DL model, from the U-Net-based architectures, imagined by us. Secondly, we will validate the performance of the DL model through a randomized controlled clinical trial. The DL model designed by us will be a highly automated tool for diagnosing and staging precancers and cervical cancer and thyroid cancers. The combined model we designed takes into account the key features of each of the architectures Overcomplete Convolutional Network Kite-Net (Kite-Net), Attention gate mechanism is an improvement added on convolutional network architecture for fast and precise segmentation of images (Attention U-Net), Harmony Densely Connected Network-Medical image Segmentation (HarDNet-MSEG). In this regard, we will create a comprehensive computer assisted diagnostic methodology validated by a randomized controlled clinical trial. The model will be a highly automated tool for diagnosing and staging precancers and cervical cancer and thyroid cancers. This would help drastically minimize the time and effort that specialists put into analyzing medical images, help to achieve a better therapeutic plan, and can provide a “second opinion” of computer assisted diagnosis.

基于FinBERT-CNN的股吧评论情感分析方法

我国股市波动受投资者情绪变化影响较大,通过对股吧等金融交流平台上投资者的评论进行情感分析,能够帮助投资者更好地了解股票市场的变化。现有的情感分析方法是利用模型对股票评论集进行分析,但缺少优质的股票评论标注数据集用于模型训练,且单一模型提取股票评论特征较为片面,模型的准确性有待提高。该文针对股吧平台上的评论数据,提出一种基于FinBERT-CNN的股吧评论情感分析方法,该方法通过FinBERT预训练模型学习股吧评论数据语义特征,解决缺乏股吧评论标注数据集的问题,并利用卷积神经网络学习股吧评论的局部特征,使模型充分学习股吧评论特征,提高模型情感分类的准确性。实验结果表明,基于FinBERT-CNN的股吧评论情感分析方法均优于现有情感分析方法。此外,通过基于股吧评论情感的股票市场关联分析,验证了股吧评论情感变化与股市波动存在相关性。

A Conceptual and Computational Framework for Aspect-Based Collaborative Filtering Recommender Systems

Many datasets in E-commerce have rich information about items and users who purchase or rate them. This information can enable advanced machine learning algorithms to extract and assign user sentiments to various aspects of the items thus leading to more sophisticated and justifiable recommendations. However, most Collaborative Filtering (CF) techniques rely mainly on the overall preferences of users toward items only. And there is lack of conceptual and computational framework that enables an understandable aspect-based AI approach to recommending items to users. In this paper, we propose concepts and computational tools that can sharpen the logic of recommendations and that rely on users’ sentiments along various aspects of items. These concepts include: The sentiment of a user towards a specific aspect of a specific item, the emphasis that a given user places on a specific aspect in general, the popularity and controversy of an aspect among groups of users, clusters of users emphasizing a given aspect, clusters of items that are popular among a group of users and so forth. The framework introduced in this study is developed in terms of user emphasis, aspect popularity, aspect controversy, and users and items similarity. Towards this end, we introduce the Aspect-Based Collaborative Filtering Toolbox (ABCFT), where the tools are all developed based on the three-index sentiment tensor with the indices being the user, item, and aspect. The toolbox computes solutions to the questions alluded to above. We illustrate the methodology using a hotel review dataset having around 6000 users, 400 hotels and 6 aspects.

Long Pulse Operation of High Performance Plasmas in JT-60U

Recent experimental progress in JT-60U advanced tokamak research is presented： sustainment of the normalized beta （βN）- 3 in a normal magnetic shear plasma, the bootstrap current fraction （fBs） - 45% in a weak shear plasma and - 75% in a reversed magnetic shear plasma in a nearly fully non-inductive current drive condition for longer than the current relaxation time. Achievement of high-density, high-radiation fraction together with high-confinement in advanced plasmas is demonstrated. Achievements and findings in long pulse operations after system modification are presented as well. A 65 s discharge of Ip = 0.7 MA was successfully obtained. As a result, high-βN of 2.3 was successfully sustained for a very long period of 22.3 s. In addition, a 30 s standard ELMy H-mode plasma of Ip up to 1.4 MA was also obtained. Effectiveness of divertor pumping to control particle recycling and the electron density under the saturated wall retention was demonstrated. These achievements and issues in development are discussed.

Potential Probiotic Tiger Nut-Cashew Nut-Milk Production by Fermentation with Two Lactic Bacteria Isolated from Ivorian Staple Foods

Probiotification of plant milk can improve its sensory and health-promoting properties. As traditional fermented foods where lactic acid bacteria (LAB) are present have been associated with beneficial effects on human health, the beneficial effects of two LAB recently isolated from two current Ivorian staple foods (a pepper and a traditional beer) were screened. These two strains LAC 1 (Lactobacillus plantarum) and LAC 2 (Pediococcus acidilactici) which presented probiotic, exopolysaccharides, inflammatory and anti-oxidant activities, were used to ferment a composite plant milk of tiger-nut and cashew (80/20) compared to two starters of a commercial yogourt. The obtained plant milks SCT 2 and SCT 3 with a significant increase in their antioxidant and/or anti-inflammatory activities and lactic bacteria contents were more preferred by consumers than SCT 1 obtained by fermentation of the commercial yogourt starters. The mixing of LAC 1 and LAC 2 was not beneficial. SCT 2 (with an anti-inflammatory activity of 31.38% and an anti-oxidant activity of 17.30%) and SCT 3 (with an anti-oxidant activity of 22.28) could be further tested in animal models to verify their nutrition-health claims.

Prevention of Gastrointestinal Pathologies: Comparative Study of the Microbial Flora of the Sanitary Surfaces of the Toilets of Students and Staff of the Félix Houphouët-Boigny University

Pathologies transmissible by hand such as gastrointestinal pathologies constitute a real public health problem, especially in sub-Saharan Africa where hygienic conditions are precarious. This study took place at Félix Houphouët-Boigny University from April to August 2018. The samples were taken from toilet surfaces such as doorknobs, tap heads, flush push buttons and seats WC. A total of three hundred and sixty-eight (368) samples were obtained, including 170 from the staff toilets and 198 from the student toilets. The results revealed the presence of total coliforms, Escherichia spp and Salmonella spp. The surfaces of student toilets were the most contaminated surfaces. The presence of entero-bacteria on the contact surfaces of the toilets of the Félix Houphouët-Boigny university represents a health risk for the university population.

Improving the Quality of Rubber Plants in Bagged Nurseries by the Use of Compost-Based Culture Substrates

The decrease of fertility soils, the rarity of quality potting soil used to fill nursery bags, the high cost of chemical fertilisers and the problems associated with their use are leading planters and rubber nurserymen in developing and/or expanding areas to look for alternative and sustainable fertilization. In this perspective, a trial was carried out at Research Station of CNRA-Bimbresso and in a farmer’s environment in order to evaluate the agronomic quality of compost made from chicken droppings and dry Panicum maximum straw to improve the growth of rubber plants in bagged nurseries. Mixtures based on potting soil and/or compost in different proportions were prepared. The follow-up of the trial focused on determination of the physico-chemical characteristics of the soils, measurement of the parameters of vegetative growth and the grafting success rate evaluation of rubber plants in nursery. The results obtained show that compost-based crop substrates increase soil organic matter, nitrogen, exchangeable bases, etc., in proportion to the doses applied. For the pH, the application of compost resulted in a reduction in soil acidity of about 1.3 unit compared to the initial values at the two study sites. The vegetative behaviour of the rubber plants also shows that qualitative (height and collar diameter) and quantitative (grafting success rate) improvements were recorded in the plants raised in compost-based substrates. The production of rubber plants in bagged nurseries was optimal with compost doses of 27 t⋅ha−1 (at Bimbresso, in the southeast) and 27 t⋅ha−1 combined with fractionated application of urea (at Kimoukro, in the centre), which under the conditions of the present study may be the recommended doses on rubber plants in bagged nurseries at Côte d’Ivoire.

成纤维生长因子受体1通过控制微小RNA let-7b表达调节内皮细胞线粒体生源的作用研究

目的探索成纤维生长因子受体l（fihrohlast growth factor receptor1，FGFRl）在调控内皮稳态时对上游分子N-乙酰基-丝氨酰-天冬氨酰-赖氨酰-脯氨酸（N-acetyl-seryl-aspartyl-lysyl-proline、AcSDKP）的影响和下游微小RNAlet-7（microRNA let-7、miR let-7）的诱导作用及与内皮细胞线粒体生源的关系，以期最终研究FGFR1在内皮稳态中的调控作用。方法下调FGF／FGFRl信号通路，蛋白质印迹（Western印迹．WB）法和细胞免疫荧光（Immunofluorescence staining）技术检测线粒体融合蛋白MFN2（mitofusin-2）、OPA1（optic atrophy protein1）和分裂蛋白DRP1（dynamin-related protein-1）的表达；应用线粒体染色（MitoTraker Green）检测线粒体的生成；运用实时荧光定量核酸扩增（Real-time Quantitative PCR、qPCR）检测技术检测microRNA let-7的表达。结果FGFRl通过诱导microRNA let-7b一5p的生成促进AcSDKP维持线粒体的生源；AcSDKP修复细胞因子复合物Triple（TGF-β2、IL-1β和TNF-α）所引起microRNA let-7b-5p表达抑制与线粒体的生成密切相关；抑制microRNA let-7b-5p的表达后，AcSDKP失去促进线粒体生成的作用：上调microRNA let-7b-5p的表达后，可修复内皮细胞中因FGF／FGFRl底物FRS2（fibroblast growth factor receptor substrate）沉默所致通路缺失所致的线粒体生成障碍。结论FGFRl通过上调microRNA let-7b-5p的表达，促进线粒体的生成，从而在内皮细胞的稳态中起着十分重要的作用。

基于t-SNE降维的科学基金资助项目可视化方法研究

【目的】设计主题模型结合流形学习文本特征降维可视化方案,更有效地发现与更直观地展示科研基金资助布局。【方法】基于美国NSF信息与智能系统(IIS)10年(2008–2017)的基金资助项目数据,利用聚类算法结合人工判读构建项目主题标签;利用TF-IDF向量空间模型与LSA潜在语义分析主题模型分别构建项目申请书高维特征,采用流形学习中t-SNE非线性降维算法将高维特征映射到二维或三维空间中可视化展示;基于构建的项目主题标签结合人工判读检验可视化效果。【结果】实验结果表明,t-SNE算法结合潜在语义分析模型在实验数据降维效果明显,可视化图谱不论在二维还是三维空间中,相同主题项目有较好的聚集性,主题间同样显示了清晰的轮廓和分界。【局限】算法参数的预设与调整需人工参与,未对不同资助机构的基金文本数据的适用性进行验证。【结论】该方法是可行的且可视化图谱能够直观地反映资助机构的资助布局,对科研管理与决策者审视宏观科研布局能够起到辅助作用。

川西铜矿周边土壤铬含量高光谱反演研究

矿产资源的开采,给其周边土壤带来许多环境问题,如何快速筛查出矿区周边土壤污染情况尤为重要。本文以川西某铜矿周边土壤作为研究对象,采用LASSO算法对SG平滑后的高光谱数据R SG和经过多元散射校正(MSC)、一阶微分(FD)、倒数变换(RT)的光谱数据进行特征波段的筛选。使用偏最小二乘回归(PLSR)、随机森林(RF)、支持向量机回归(SVR)、极端梯度提升(XGBoost)、反向传播神经网络(BPNN)5种模型进行反演。结果表明,R SG、MSC、FD筛选出的特征波段集中在近红外区域,RT筛选出的特征波段集中在可见光区域;MSC-SVR模型的R 2,RMSE和RPD分别为0.763、6.745、2.06,在所有模型中精度最高,该模型可用于研究区土壤中铬的快速监测。

松材线虫胁迫对5种松树的代谢组共性表达

为研究松树在松材线虫胁迫下的分子防御机制,分别选取15~20年生的辽宁抚顺的油松(LY)、辽宁抚顺的红松(LH)、山东威海的赤松(SC)、山东威海的黑松(SH)、湖南长沙的马尾松(HM)5种代表性松树作为研究对象,运用液相色谱-质谱联用技术的非靶向代谢组学,将以上5种松树分为健康组和感病组,分别对5组松树样本的差异代谢物分布情况、差异代谢物的共性特征、共有差异代谢物的表达显著性、代谢通路富集分布情况、差异性代谢通路的共性特征等进行研究。结果表明:5种松树感染松材线虫后的代谢产物在苯和取代衍生物、羧酸及其衍生物、脂肪酰基、酚类、孕烯醇酮脂类、类固醇及其衍生物、其他类等7类化合物中分布较为集中;HM、LH、LY和SC,HM、LY、SC和SH这两种组合,只有1种共有的显著性差异代谢物,分别是百里酚和6-姜酮酚,5种松树以及剩下的其他4种松树的组合都没有共有的显著性差异代谢物;HM、LY、SC和SH感染松材线虫后,6-姜酮酚表达量显著升高,LH感染松材线虫后,6-姜酮酚表达量无显著性差异;HM、SC、LY和LH感染松材线虫后,百里酚表达量显著升高,SC感染松材线虫后,百里酚表达量无显著性差异;HM、LH、LY、SC、SH分别富集到51、55、46、71、43条差异显著的信号通路,5种松树共有的差异显著的信号通路共有19条,主要反映在植物的抗逆相关信号通路上,包括花生四烯酸代谢、亚油酸代谢、苯丙烷生物合成、多种氨基酸代谢等。

技术主题动态演化分析的一种新方法:DPL-BMM模型

揭示技术演化脉络是把握技术发展规律的前提,基于专利信息的主题挖掘是基于技术发展微观机制呈现宏观规律的重要研究内容,对技术超前布局和创新驱动实践具有重大意义。技术主题动态演化分析DPL-BMM(Dirichlet process biterm-based mixture model with labelling)是一种附有标签的基于双项狄利克雷过程的混合模型,其突破了传统主题模型在进行主题识别时需固定主题数目的局限,通过增加技术主题表示模块使识别到的技术主题内容更加明确。本文以人工智能领域技术为例进行实证分析,研究结果表明,该方法对技术主题及其演化脉络展示具有实际应用价值。

松辽盆地西缘构造带构造格架、演化及油气资源潜力

松辽盆地西缘构造带是研究松辽盆地形成演化的关键地带,但目前关于该构造带的构造格架、活动历史、控盆特征等仍不清晰,构造带深部物质架构及中—新生代构造演化的精细厘定为解决这一科学问题提供了新的证据。松辽盆地西缘地区布格重力异常与综合地球物理剖面特征指示,松辽盆地西缘构造带为区域上带状展布的超壳深大构造带。通过对综合构造变形带的产状、性质、运动学等资料分析,提出松辽盆地西缘构造带东西分带、倾向相背的“两带八字形”构造格架。依据构造带发育的基底背景、构造样式、地球物理特征等差异,建立了南北分段的“四段式”展布构架。在此基础上,结合区域构造演化、变形带构造解析与40Ar/39Ar构造年代学等资料,恢复了晚三叠世、中晚侏罗世、早白垩世中期、晚白垩世中期、古新世—渐新世和新近纪—更新世6期中—新生代构造活动年代格架。结合构造带的控盆作用指出,外盆带控制的中侏罗世和早白垩世中期沉积盆地是油气资源勘查的重要选区,为中国东北地区新层系油气资源勘查提供了新依据。