摘要：【中文】本公开涉及一种故障相关信息辅助检索方法、装置、存储介质及电子设备。方法包括：依次将工业知识图谱中的各节点作为目标节点，若工业知识图谱中存在指向目标节点的其他节点，则根据其他节点的目标检索热度值对目标节点的第一检索热度值进行修正；响应于接收到故障检索页面的访问请求，根据工业知识图谱中各节点的第二检索热度值，确定预设数量的热门故障相关信息并显示之。由此，可以提升热门故障相关信息的准确性，使得热门故障相关信息中包含用户当前实际所需检索的故障相关信息的概率增大。相应地，用户大多可以直接通过点选操作获取检索结果，而无需手动输入检索信息进行检索，方便快捷，从而提升了用户体验。 【EN】The disclosure relates to a fault-related information auxiliary retrieval method and device, a storage medium and an electronic device. The method comprises the following steps: sequentially taking each node in the industrial knowledge graph as a target node, and if other nodes pointing to the target node exist in the industrial knowledge graph, correcting a first retrieval heat value of the target node according to the target retrieval heat values of the other nodes; and responding to the received access request of the fault retrieval page, and determining and displaying a preset amount of hot fault related information according to the second retrieval heat value of each node in the industrial knowledge graph. Therefore, the accuracy of the hot fault related information can be improved, and the probability that the hot fault related information contains the fault related information which is actually required to be searched by the user at present is increased. Correspondingly, most users can directly obtain the retrieval result through clicking operation without manually inputting retrieval information for retrieval, and the method is convenient and fast, so that the user experience is improved.

摘要：【中文】本发明实施例公开了一种产品数据的查看方法、装置、设备及存储介质。其中，该方法包括：确定用户选择的多个理财产品；根据所述多个理财产品的产品标签和用户标签，生成所述多个理财产品的对比标签集合；根据用户从所述对比标签集合中选择的标签，生成所述多个理财产品的对比数据，以供用户查看。本发明实施例解决了现有技术中，用户只能在选择购买产品前进行产品对比和手动提取标签数据的问题，有效提高了产品数据的查询效率。 【EN】The embodiment of the invention discloses a method, a device, equipment and a storage medium for checking product data. Wherein, the method comprises the following steps: determining a plurality of financial products selected by a user; generating a comparison label set of the plurality of financial products according to the product labels and the user labels of the plurality of financial products; and generating comparison data of the plurality of financial products according to the label selected by the user from the comparison label set so as to be viewed by the user. The embodiment of the invention solves the problems that in the prior art, a user can only compare products and manually extract tag data before selecting to purchase the products, and effectively improves the query efficiency of product data.

摘要：【中文】本发明提供了一种协同任务预测方法及计算机可读存储介质；该方法包括：接收多个终端发送的样本身份信息，基于样本身份信息，通过与多个终端进行中间结果交互所得到的子纵向模型，构造出纵向模型；接收多个终端发送的多个子横向模型，并对多个子横向模型进行整合，得到横向模型；对纵向模型和横向模型进行整合，得到任务预测模型，并将任务预测模型发送给多个终端；获取待预测身份信息，并将待预测身份信息分别发送给多个终端；当接收到多个终端针对待预测身份信息所返回的多个子预测结果时，利用多个子预测结果整合出任务预测结果，完成协同任务预测。通过本发明，能够提高任务预测的准确度。 【EN】The invention provides a cooperative task prediction method and a computer readable storage medium; the method comprises the following steps: receiving sample identity information sent by a plurality of terminals, and constructing a longitudinal model through a sub-longitudinal model obtained by performing intermediate result interaction with the plurality of terminals based on the sample identity information; receiving a plurality of sub-transverse models sent by a plurality of terminals, and integrating the plurality of sub-transverse models to obtain a transverse model; integrating the longitudinal model and the transverse model to obtain a task prediction model, and sending the task prediction model to a plurality of terminals; acquiring identity information to be predicted, and respectively sending the identity information to be predicted to a plurality of terminals; and when a plurality of sub-prediction results returned by a plurality of terminals aiming at the identity information to be predicted are received, integrating the task prediction results by using the plurality of sub-prediction results to complete the collaborative task prediction. By the method and the device, the accuracy of task prediction can be improved.

摘要：【中文】本发明涉及新型喜树碱衍生物及其应用、肿瘤细胞生长抑制剂和三元复合物以及提高喜树碱衍生物溶解性的方法。所述的喜树碱衍生物由式1所示的物质经糖基化三氮唑在R³位置修饰而成；所述式1所示的结构式中R¹代表H、C_1‑10的烷基、C_1‑10的氘代烷基或C₁‑C₁₀的卤代烷基；R²代表H、CH₂N(CH₃)₂或CH₂N(CD₃)₂；R⁴代表或H，其中X代表N，O或S；L代表多肽、C_1‑20直链烷基或其衍生物、C_1‑20直链或支链酰基衍生物、C_2‑100乙二醇或其衍生物。所述喜树碱衍生物溶解度较高，制备的抗癌药物具有抗癌谱广、安全性高的优点，体内抗癌活性优于盐酸伊立替康。 【EN】The invention relates to novel camptothecin derivatives, applications thereof, tumor cell growth inhibitors and ternary complexes, and a method for improving the solubility of camptothecin derivatives. The camptothecin derivative is prepared by subjecting a substance shown as a formula 1 to glycosylation triazole at R³Position modification; r in the structural formula shown in the formula 1¹Representative H, C_1‑10Alkyl of (C)_1‑10Deuterated alkyl or C₁‑C₁₀A haloalkyl group of (a); r²Representative H, CH₂N(CH₃)₂Or CH₂N(CD₃)₂；R⁴RepresentsOr H, wherein X represents N, O or S; l represents a polypeptide, C_1‑20Straight-chain alkyl or derivative thereof, C_1‑20Straight or branched acyl derivative, C_2‑100Ethylene glycol or a derivative thereof. The camptothecin derivative has high solubility, and the prepared anticancer drug has the advantages of wide anticancer spectrum and high safety, and the in vivo anticancer activity of the camptothecin derivative is superior to that of irinotecan hydrochloride.

摘要：【中文】一种基于欺骗攻击者的对抗样本防御方法,属于计算机图像处理技术领域。技术方案：在不改变原始分类器分类性能的基础上构建分类系统，采用并行的方式将缓冲器添加到分类系统中，通过构造与原始分类器泛化性能较差的分类器2，很好地欺骗攻击者；在原始分类器之前设置额外的检测器，防御单步攻击以及提高攻击过程的难度，只采用原始分类器对被检测器分类为正常标签的图像进行分类；采用比较器或集成学习块更换检测器，获得更好的分类性能以及无梯度回传的特性，比较器不承担任何特定的攻击策略和扰动模式。有益效果是:本发明采用缓冲器、比较器等阻碍了对抗样本的生成过程，在不影响正常样本的分类性能下，提升了分类网络对对抗样本的防御性能。 【EN】A defense method for confrontation samples based on deception attackers belongs to the technical field of computer image processing. The technical scheme is as follows: constructing a classification system on the basis of not changing the classification performance of an original classifier, adding a buffer into the classification system in a parallel mode, and well deceiving an attacker by constructing a classifier 2 with poor generalization performance with the original classifier; an additional detector is arranged in front of an original classifier, single-step attack is prevented, the difficulty of an attack process is improved, and only the original classifier is adopted to classify the images classified into normal labels by the detector; the comparator or the integrated learning block is adopted to replace the detector, so that better classification performance and the characteristic of no gradient return are obtained, and the comparator does not bear any specific attack strategy and disturbance mode. The method has the advantages that the generation process of the countermeasure sample is hindered by adopting the buffer, the comparator and the like, and the defense performance of the classification network on the countermeasure sample is improved under the condition that the classification performance of the normal sample is not influenced.

摘要：【中文】本发明提供一种建筑测量机器人多传感器融合三维建模方法，通过双目结构光深度相机与多线扫描激光雷达在水平和垂直方向上的旋转，完成单站点360度全景扫描；融合单线激光扫描雷达、车轮里程计以及惯性测量单元IMU等多传感器信息，自动输出导航定位信息，完成多测量站点间三维点云的初始配准；融合RGB图像与三维点云，完成关键特征点的成对点对几何特征的提取与匹配，实现多站点间三维点云的精确拼接。本发明还提供一种建筑测量机器人多传感器融合三维建模系统。本发明解决在建筑工地等复杂环境下，基于多传感器融合，针对纹理和几何结构简单的室内场景，进行精确的三维建模的问题。 【EN】The invention provides a building surveying robot multi-sensor fusion three-dimensional modeling method, which completes single-site 360-degree panoramic scanning through the rotation of a binocular structured light depth camera and a multi-line scanning laser radar in the horizontal and vertical directions; multi-sensor information such as a single-line laser scanning radar, a wheel odometer and an inertial measurement unit IMU is fused, navigation positioning information is automatically output, and initial registration of three-dimensional point cloud among multiple measurement stations is completed; and fusing the RGB image and the three-dimensional point cloud, completing the extraction and matching of the paired point pair geometric features of the key feature points, and realizing the accurate splicing of the three-dimensional point cloud among multiple stations. The invention also provides a multi-sensor fusion three-dimensional modeling system of the building surveying robot. The invention solves the problem of accurate three-dimensional modeling aiming at indoor scenes with simple textures and geometric structures based on multi-sensor fusion in complex environments such as construction sites and the like.

摘要：【中文】一种基于权重谱生成对抗样本的点攻击方法,属于计算机图像处理技术领域。技术方案：计算奇异点及其邻近点(4或8个点)之间的平均距离作为评估指标区分包含奇异点的对抗样本；设计优化函数来提高抗检测性能，修改优化步骤以实现基于点的攻击，解决显著性谱存在的问题,最终实现设计新的基于点的攻击方法生成抗感知能力以及抗检测性能较好的对抗样本；选择直接对不稳定区域(权重较大的区域)设置更多的篡改，为方便权衡篡改点的数量、篡改程度以及攻击成功率引入新的约束。有益效果是:本发明用生成的权重谱决定需要篡改点的位置以及篡改程度，同时在指定篡改程度的前提下，可以采用固定点数或最少像素点攻击方法生成对抗样本。 【EN】A point attack method for generating an anti-sample based on a weight spectrum belongs to the technical field of computer image processing. The technical scheme is as follows: calculating the average distance between the singular point and the adjacent points (4 or 8 points) thereof as an evaluation index to distinguish the confrontation samples containing the singular point; designing an optimization function to improve the detection resistance, modifying the optimization step to realize point-based attack, solving the problems of the significance spectrum, and finally realizing the generation of a new point-based attack method to generate an antagonistic sample with better perception resistance and detection resistance; and directly setting more falsifications to unstable areas (areas with higher weight), and introducing new constraints for conveniently weighing the number of falsification points, the falsification degree and the attack success rate. The method has the advantages that the generated weight spectrum is used for determining the position of the tampering point and the tampering degree, and meanwhile, on the premise of specifying the tampering degree, the countermeasure sample can be generated by adopting a fixed point number or minimum pixel point attacking method.

摘要：【中文】本发明提供了一种冰箱杀菌装置，包括：容器，设置于冰箱内，其外壳内形成有内腔，其中所述外壳上具有多个第一孔；噬菌体溶液，容纳于所述内腔内，且所述噬菌体溶液自所述第一孔排出所述容器，并进入所述冰箱的储物间室。本发明的冰箱杀菌装置利用噬菌体来杀菌，杀菌效果好，且噬菌体能特异性识别细菌真菌，对食物和人体没有影响，同时整个冰箱杀菌装置的结构简单，易于配置，不会产生异味，用户体验好。本发明还提供一种具有该冰箱杀菌装置的冰箱。 【EN】The invention provides a refrigerator sterilization device, comprising: the refrigerator comprises a container, a first cover and a second cover, wherein the container is arranged in the refrigerator, an inner cavity is formed in a shell of the container, and a plurality of first holes are formed in the shell; and the phage solution is contained in the inner cavity, is discharged out of the container from the first hole and enters the storage chamber of the refrigerator. The refrigerator sterilization device provided by the invention utilizes the phage for sterilization, the sterilization effect is good, the phage can specifically identify the bacteria and fungi, no influence is caused on food and human bodies, and meanwhile, the whole refrigerator sterilization device is simple in structure, easy to configure, free of peculiar smell and good in user experience. The invention also provides a refrigerator with the refrigerator sterilization device.

摘要：【中文】本发明公开了一种依存关系对齐组件、依存关系对齐训练方法、设备及介质，所述的依存关系对齐组件包括：词对齐信息接收模块，用于接收第一语言语料和第二语言语料各自的多个翻译结果以及对应的多个词对齐关系信息；依存树信息接收模块，用于接收所述第一语言语料和所述第二语言语料各自的多个句法依存树；对齐筛选模块，用于对所述多个翻译结果的对齐程度和所述多个句法依存树的对齐程度进行筛选。利用本发明实施例可获得双语句法对齐的高质量树库。 【EN】The invention discloses a dependency relationship alignment component, a dependency relationship alignment training method, a device and a medium, wherein the dependency relationship alignment component comprises: the word alignment information receiving module is used for receiving a plurality of translation results of the first language corpus and the second language corpus and a plurality of corresponding word alignment relation information; a dependency tree information receiving module, configured to receive a plurality of syntax dependency trees of the first language corpus and the second language corpus, respectively; and the alignment screening module is used for screening the alignment degrees of the translation results and the alignment degrees of the syntactic dependency trees. The embodiment of the invention can obtain the bilingual syntax aligned high-quality tree library.

摘要：【中文】本申请公开了一种套管旋转装置，包括：支架；转动的设置在支架的顶部，并能够固定连接套管的连接部件；设置在支架顶部的驱动部件，该驱动部件能够驱动连接部件及套管进行转动；与驱动部件电连接，并控制其工作的控制器。上述的套管旋转装置，可以使套管在分别转动至多个不同角度时，都对其进行一次试验操作，从而得到套管处于不同角度的更加全面的试验数据，更加真实的模拟了套管实际运行时的安装角度，令试验结果的精确度得到提升。并且支架不再是临时布置，而是一个完整的成型结构，当需要对套管进行试验，则可以直接使用该支架，令套管的试验周期得到了缩短、试验效率得到了提升。本发明还提供了适用于上述套管旋转装置的一种套管试验方法。 【EN】The application discloses sleeve pipe rotary device includes: a support; the connecting part is rotatably arranged at the top of the bracket and can be fixedly connected with the sleeve; the driving part is arranged at the top of the bracket and can drive the connecting part and the sleeve to rotate; and a controller electrically connected to the driving part and controlling the operation thereof. The sleeve rotating device can perform one-time test operation on the sleeve when the sleeve rotates to a plurality of different angles respectively, so that more comprehensive test data of the sleeve at different angles can be obtained, the installation angle of the sleeve during actual operation can be simulated more truly, and the accuracy of a test result can be improved. And the support is not temporarily arranged any more, but is a complete molding structure, when the sleeve is required to be tested, the support can be directly used, so that the test period of the sleeve is shortened, and the test efficiency is improved. The invention also provides a sleeve test method suitable for the sleeve rotating device.