Datasets:

CHINAYWX

/

g06f

like 0

接收单元，接收所述回应信息；

The receiving unit receives the response information;

储存单元，用以储存多媒体信息；

Storage unit for storing multimedia information;

以及处理单元，耦接所述传送单元、所述接收单元与所述储存单元，用以产生所述浏览动作的指令与解析所述回应信息，以产生所述多媒体信息；

The processing unit is coupled with the transmission unit, the receiving unit and the storage unit to generate the instruction of the browsing action and analyze the response information to generate the multimedia information;

其中，所述处理单元获取针对所述多媒体服务器的根节点之下的前面部分多个第一层节点的底下的具有最多子节点数的第二层节点，且所述处理单元指示所述储存单元将所述具有最多子节点数的多个第二层节点的至少一部分的物件识别与子节点数作为所述多媒体信息的一部分。

Wherein, the processing unit obtains a second-layer node with the maximum number of child nodes under the previous part of a plurality of first-layer nodes under the root node of the multimedia server.

针对所述前面部分多个第一层节点的每一者，所述处理单元用以获得所述第一层节点底下的所述前面部分多个第二层节点中具有最多子节点数的第二层节点的物件识别与子节点数；

For each of the first layer nodes of the preceding part, the processing unit is used to obtain the object identification and the number of child nodes of the second layer node with the largest number of child nodes among the first layer nodes of the preceding part under the first layer node;

针对具有最多子节点数的多个第二层节点的每一者，若所述第二层节点底下其中一个子节点为多媒体文件，则所述处理单元将指示所述储存单元记录所述具有最多子节点数的第二层节点的物件识别与子节点数为所述多媒体信息的一部分。

For each of a plurality of Layer 2 nodes with a maximum number of child nodes, if one of the child nodes under the Layer 2 node is a multimedia file, the processing unit will instruct the storage unit to record the object identification of the Layer 2 node with the maximum number of child nodes and the number of child nodes as part of the multimedia information.

以及至少一多媒体服务器，连结所述多媒体装置；

And at least one multimedia server connecting the multimedia device;

所述多媒体装置用以获得所述多媒体服务器的根节点底下的前面部分多个第一层节点的物件识别；

The multimedia device is used to obtain object identification of a plurality of first-layer nodes in the preceding part under the root node of the multimedia server;

针对所述前面部分多个第一层节点的每一者，所述多媒体装置用以获得所述第一层节点底下的部分多个第二层节点中具有最多子节点数的第二层节点的物件识别与子节点数；

For each of the first layer nodes in the preceding part, the multimedia device is used to obtain the object identification and the number of child nodes of the second layer node with the largest number of child nodes among the first layer nodes under the first layer node;

针对具有最多子节点数的多个第二层节点的每一者，若所述第二层节点底下其中一个子节点为多媒体文件，则所述多媒体装置记录所述第二层节点的物件识别与子节点数为多媒体信息的一部分。

For each of a plurality of Layer 2 nodes having a maximum number of child nodes, if one of the child nodes under the Layer 2 node is a multimedia file, the multimedia device records the object identification of the Layer 2 node and the number of child nodes as part of the multimedia information.

针对所述前面部分多个第一层节点的每一者，获得所述第一层节点底下的前面部分多个第二层节点中具有最多子节点数的第二层节点的物件识别与子节点数；

For each of the first layer nodes in the previous part, the object identification and child node number of the second layer node with the largest number of child nodes among the first layer nodes in the previous part under the first layer node are obtained;

以及针对具有最多子节点数的多个第二层节点的每一者，若所述第二层节点底下其中一个子节点为多媒体文件，则记录所述第二层节点的物件识别与子节点数为多媒体信息的一部分。

And for each of a plurality of Layer 2 nodes with a maximum number of child nodes, if one of the child nodes under the Layer 2 node is a multimedia file, the object identification of the Layer 2 node and the number of child nodes are recorded as part of the multimedia information.

以及若所记录的多个第二层节点底下的子节点中有两者的类别彼此相同，则所述多媒体装置检查两者类别相同的第二层节点底下的部分多个子节点的类别，以将所述第二层节点底下的部分多个子节点的类别彼此不同的第二层节点的物物件识别与子节点数自多媒体信息删除。

And if the categories of two sub-nodes under the recorded multiple second-layer nodes are the same with each other, the multimedia device checks the categories of some sub-nodes under the second-layer nodes with the same categories to delete the object identification and the number of sub-nodes of the second-layer nodes with different categories from each other.

一种多类别集工业过程数据相似度多维度分析方法，具体步骤为：1)设定类别集；

A multi-class set multi-dimensional analysis method for industrial process data similarity, the specific steps are as follows: 1) set the class set;

本发明针对不同的数据维度，对部分数据特征进行相似度分析，从而实现对不同类别集的数据进行相似度分析，来找出各类别集之间存在的隐含联系，研究系统中产生不同运行状态的原因，以便于数据的挖掘、检索与分析，寻找工况产生的原因。

Aiming at different data dimensions, the invention carries out similarity analysis on some data features, so as to realize similarity analysis on different sets of data, find out the hidden relationship between various sets, study the causes of different operating states in the system, so as to facilitate data mining, retrieval and analysis, and find the causes of working conditions.

本发明公开了一种分布式数据同步到数据仓库的方法及装置，用以解决现有技术提供的方案中，数据仓库中表的维护数量和难度较大，同步流程复杂，耗费时间长，可靠性低的问题。

The invention discloses a method and device for synchronizing distributed data to a data warehouse, which solves the problems of large maintenance quantity and difficulty of tables in the data warehouse, complex synchronization process, long time consuming and low reliability in the scheme provided by the prior art.

分布式数据同步到数据仓库的方法，包括：将分布式数据库的每一个分表中的增量数据导出为文本，并对应生成文本文件；

The method of synchronizing distributed data to the data warehouse includes: exporting the incremental data in each sub-table of the distributed database as text, and generating the corresponding text file;

将各文本文件中的文本导入到数据仓库的增量表中；

Import text from each text file into the incremental table of the data warehouse;

根据特定的分表规则将增量表中的全部增量数据合并到数据仓库的汇总表中，所述分表规则是指将集中式数据库的大表中存储的全部数据划分到分布式数据库的多个分表中时所遵循的规则。

All the incremental data in the incremental table is merged into the summary table of the data warehouse according to a specific subtable rule, which refers to the rules that are followed when all the data stored in the large table of the centralized database is divided into multiple subtables of the distributed database.

一种用于从数据系统的数据库中选择待同步的数据集合的方法，在该方法中，存储说明数据系统中数据单元之间关系的系统元数据、以便选择待同步的数据集合。

A method for selecting a data set to be synchronized from a database of a data system in which system metadata describing the relationships between data units in a data system is stored in order to select a data set to be synchronized.

所述元数据至少包括关于数据单元之间相关性的信息。

The metadata includes, at a minimum, information about correlations between data units.

当第一数据集合被同步时，检索与至少一个第一数据集合中的初始数据单元相关联的元数据。

When the first data set is synchronized, the metadata associated with the initial data unit in at least one of the first data sets is retrieved.

接着，选择用于同步的第二数据集合，所述第二数据集合根据至少一个元数据而包括具有相对于所述初始数据单元的最大相关性值的数据单元。

Next, select a second data set for synchronization that includes a data unit with a maximum correlation value relative to the initial data unit based on at least one piece of metadata.

按照预先定义的主题类别，为网页文档集合的每个网页文档确定一个类别；

Identify a category for each Web document in the web document collection based on a predefined topic category;

对每个类进行访问频度统计，访问频度值最高的网页集作为用户感兴趣的相关网页，作为分析数据，进而实现数据业务的定向推送，提高数据业务推送的可信度。

The access frequency statistics of each class are carried out. The set of web pages with the highest access frequency is taken as relevant web pages interested by users and used as analysis data to achieve targeted data service push and improve the credibility of data service push.

本发明提供了一种控制网络链接快捷方式分类的方法和系统，所述方法包括：获取网络链接快捷方式的网络地址；

The invention provides a method and system for controlling the classification of a network link shortcut. The method comprises: obtaining a network address of a network link shortcut;

根据所述网络地址识别所述网络链接快捷方式所属的类别，并生成所述网络链接快捷方式在所述类别所属的文件夹目录信息；

Identify the category to which the network link shortcut belongs according to the network address, and generate the folder directory information to which the network link shortcut belongs in the category;

按照所述文件夹目录信息在浏览器中进行所述网络链接快捷方式的同步。

Synchronize the network link shortcut in the browser according to the folder directory information.

所述系统包括：地址获取模块、识别模块和同步模块。

The system comprises an address acquisition module, an identification module and a synchronization module.

采用本发明能提高操作上的方便性。

The convenience of operation can be improved by adopting the invention.

上传所述提取的网络地址。

Upload the extracted network address.

创建所述网络链接快捷方式在所述类别所属的文件夹目录信息。

Creates the folder directory information to which the network link shortcut belongs in the category.

判断所述获取的目录信息中是否存在所述提取的类别所属的文件夹目录信息，若是，则在所述提取的类别所属的文件夹目录信息中添加所述网络链接快捷方式的网络地址，若否，则创建所述提取的类别所属的文件夹目录信息，并将所述网络链接快捷方式的网络地址添加至所述创建的文件夹目录信息。

Determine whether the folder directory information to which the extracted category belongs exists in the obtained directory information; if so, add the network address of the network link shortcut to the folder directory information to which the extracted category belongs; if not, create the folder directory information to which the extracted category belongs.

识别模块，用于根据所述网络地址识别所述网络链接快捷方式的类别，并生成所述网络链接快捷方式在所述类别所属的文件夹目录信息；

The identification module is used to identify the category of the network link shortcut according to the network address, and generate the folder directory information to which the network link shortcut belongs in the category;

上传单元，用于上传所述提取的网络地址。

Upload unit for uploading the extracted network address.

所述类别提取单元用于提取与所述标识字段相关联的类别；

The category extraction unit is used to extract a category associated with the identification field;

创建单元，用于创建所述网络链接快捷方式在所述类别所属的文件夹目录信息。

Creates a unit for creating folder directory information to which the network link shortcut belongs in the category.

判断单元，用于判断所述获取的目录信息中是否存在所述提取的类别所属的文件夹目录信息，若是，则通知添加单元，若还，则通知创建添加单元；

The judgment unit is used to judge whether the obtained directory information exists the folder directory information to which the extracted category belongs. If so, the adding unit is notified; if it is returned, the creation of adding unit is notified;

所述添加单元用于在所述提取的类别所属的文件夹目录信息中添加所述网络链接快捷方式的网络地址；

The addition unit is used to add the network address of the network link shortcut to the folder directory information to which the extracted category belongs;

所述创建添加单元用于创建所述提取的类别所属的文件夹目录信息，并将所述网络链接快捷方式的网络地址添加至所述创建的文件夹目录信息。

The create add unit is used to create the folder directory information to which the extracted category belongs and to add the network address of the network link shortcut to the created folder directory information.

本发明公开了一种电力系统广域控制系统的数据处理方法，该方法包括以下内容：广域控制系统数据传输通道随机延时实时测量技术；

The invention discloses a data processing method for a wide-area control system of a power system, which comprises the following contents: a real-time measurement technique for random delay of the data transmission channel of the wide-area control system;

广域控制系统数据传输通道随机延时实时补偿技术；

Real-time compensation technology for random delay of data transmission channel in wide area control system;

广域控制系统实时数据库技术；

Real-time database technology of wide area control system;

广域控制系统实时数据库容错技术；

Fault tolerance technology of real-time database of wide area control system;

广域控制系统数据质量标记技术。

Wide area control system data quality labeling technology.

本发明属于电力系统广域控制技术领域，本发明的技术方案用以解决电力系统广域控制系统中的数据传输通道随机延时与故障带来的数据处理难题，为电力系统广域控制系统提供了一种可靠的数据处理方法。

The invention belongs to the technical field of wide-area control of power system, and the technical scheme of the invention is used to solve the data processing problem caused by random delay and fault of the data transmission channel in the wide-area control system of power system, and provides a reliable data processing method for the wide-area control system of power system.

第二种方式为，执行站将通道延时计算结果上传给主站，由主站对数据进行预补偿后再下发给执行站，其中预补偿是根据以往通道延时的平均值或最新值完成的，执行站在收到数据后，还需要判定预补偿是否有效。

The second method is that the execution station sends the channel delay calculation result to the master station, and the master station pre-compensates the data and then sends it to the execution station. The pre-compensation is completed according to the average value or the latest value of the previous channel delay. The execution station also needs to determine whether the pre-compensation is valid after receiving the data.

实时数据库存放数据、数据的时间标签和数据的可靠性信息；

Real-time database to store data, data time label and data reliability information;

该实时数据库对同一测点的数据缓存多个时间断面；

The real-time database caches multiple time sections of data from the same measurement point.

该实时数据库采用向系统一次申请多个时间断面的内存，循环覆盖使用的策略，从而保证数据的存取效率；

The real-time database adopts the strategy of applying the memory of multiple time sections to the system at one time and using it in cycle, so as to ensure the efficiency of data access.

该实时数据库提供某一采集站的当前最新时标及查询接口；

The real-time database provides the latest time scale and query interface of a collection station.

该实时数据库还提供依据全球统一时标获取数据的查询接口。

The real-time database also provides a query interface to obtain data according to the globally unified time scale.

其二是对采集站时标异常的容错技术：采集站时标是采集的全球统一时标，出于各种原因，采集站有时无法获得满意精度的时标，此时在采集站会标记该时标不可靠，主站在计算实时库最新时标时，不可靠时标不计算在内，避免采集站的错误时标偏差太大引起实时数据库无法正常工作。

The second is the fault-tolerant technology for the time scale anomaly of the acquisition station:

广域控制系统主站实时数据库中设计了数据的存储位置、数据时标的存储位置和数据质量的存储位置；

In the real-time database of the master station of wide area control system, the storage location of data, the storage location of data time object and the storage location of data quality are designed.

当采集站上传的数据包中标记了数据不可靠时，将标记存储到实时数据库的对应位置；

When the data packet uploaded by the acquisition station is marked as unreliable, the mark is stored in the corresponding position of the real-time database;

实时数据库内存是循环覆盖重用的，更新数据同时也会刷新数据所附带的全球统一时间标签；

Real-time database memory is cyclic overwrite reuse, updating data will also refresh the global unified time label attached to the data;

当采集站数据包丢失，造成数据缺失时，由于实时数据库是内存循环覆盖重用的，相应内存位置的时间标签与实时库的整体时间标签会相差一个到多个循环周期，从而有效标记该位置数据是未更新的旧数据。

When the data packet of the acquisition station is lost, resulting in data loss, because the real-time database is overwritten and reused by memory cycles, the time label of the corresponding memory location will differ from the overall time label of the real-time library by one to several cycle cycles, thus effectively marking the location data as unupdated old data.

本发明公开了一种基于问答库的中文自然语言问答方法。

The invention discloses a Chinese natural language question answering method based on question answering database.

与现有的搜索引擎不同，可接受自然语言方式的用户查询，并且直接返回答案文本，用户无需再在结果中手动筛选；

Different from the existing search engines, it can accept user queries in natural language, and directly return the answer text, the user does not need to manually filter in the results;

同时本发明利用已有的答案来回答用户的问题，这些答案通常是手工生成，格式规整，且具有一定的权威性；

At the same time, the invention uses the existing answers to answer the user's questions, which are usually manually generated, regular format, and have a certain authority;

同时所能解决的问题也不受问题类型的限制，易于广泛应用，可在一定程度上克服原有问答系统存在的弊端。

At the same time, the problems that can be solved are not limited by the types of problems, and it is easy to be widely used, which can overcome the drawbacks of the original question answering system to a certain extent.

本发明提供了一种生物认证方法和生物认证系统。

The invention provides a biometric authentication method and a biometric authentication system.

该生物认证系统检测身体特征，针对已记录的生物数据来验证所述特征，以进行个人认证。

The biometric authentication system detects physical characteristics and verifies said characteristics against recorded biometric data for personal authentication.

用于可扩展上下文感知的自然语言交互的技术包含具有若干上下文源模块的计算装置。

Techniques for extensible context-aware natural language interactions consist of computing devices with several context source modules.

上下文源模块可在向用户部署计算装置之后开发或安装。

Context source modules can be developed or installed after a computing device is deployed to a user.

每个上下文源模块包含上下文捕获模块、语言模型、一个或多个数据库查询映射，并且可包含一个或多个用户接口元素映射。

Each context source module contains a context capture module, a language model, one or more database query mappings, and can contain one or more user interface element mappings.

上下文捕获模块解释、生成并且存储上下文数据。

The context capture module interprets, generates, and stores context data.

描述并要求保护其它实施例。

Describe and require protection of other embodiments.

本发明提供一种用于人物搜索的方法，其中，该方法包括以下步骤：获取来自用户的搜索序列；

The invention provides a method for searching people, which comprises the following steps: obtaining a search sequence from a user;

由所述搜索序列中获取至少一个搜索分词；

Obtain at least one search segmentation from the search sequence;

基于所述至少一个搜索分词获得至少一个候选人物，其中，所述至少一个搜索分词的全部或部分中的每个搜索分词均能与一个候选人物的人物标签中的一个或多个匹配；

Obtain at least one candidate character based on at least one search word, where each search word in all or part of the at least one search word matches one or more of the person tags of a candidate character;

根据各个候选人物的、与所述搜索分词匹配的人物标签的权重信息，对所述至少一个候选人物中的全部或部分候选人物进行排序；

Rank all or part of the candidates in at least one candidate according to the weight information of the person tags matching the search segmentation of each candidate;

将排序后的候选人物提供给所述用户。

Presents the sorted candidate characters to the user.

本发明公开了一种对二维表格进行特征分析和数据提取的方法，该方法可以对符合所述特征的示例表格进行特征分析和根据指定的关键进行数据提取，组成字典数据，然后根据这份数据字典对示例表进行数据更新。

The invention discloses a method for feature analysis and data extraction of a two-dimensional table. The method can perform feature analysis on a sample table conforming to the features and extract data according to the specified key, compose dictionary data, and then update the sample table according to the data dictionary.

本发明方法可以应用于将一份表格的数据按字段导入到另一份表格中的需求，比如：根据客户提供的翻译文件导入数据到手机平台上的翻译文件。

The method of the invention can be applied to the requirement of importing data from one form to another form by field, for example, importing data from the translation file provided by the customer to the translation file on the mobile phone platform.

本发明与现有技术相比不仅能够获取语法语义信息，还能从少量的带标注的文本中获得类别信息，使得用户搜索词向量在空间上面不同类别的数据分得更加开，提高了分类的准确性，有效解决了用户查询文本短的问题，方法简便，准确性好。

Compared with the prior art, the invention can not only obtain syntactic and semantic information, but also obtain category information from a small amount of marked text, so that the user's search term vector can be more separated from the data of different categories in space, improve the accuracy of classification, and effectively solve the problem of short text query by the user, with simple method and good accuracy.

所述已标注的搜索词为人工标注或规则标注的搜索词；

The marked search terms are manually marked or regularly marked search terms;

通过系统传感器捕捉手或眼睛动作的手势引导输入；

Gesture-guided input that captures hand or eye movements through system sensors;

向虚拟辅助子系统发出命令，向辅助子系统发出的该命令可为全景视图也可通过聊天方式发出。

A command is issued to the virtual auxiliary subsystem, which can be issued in panoramic view or via chat.

一个用于显示三维图形环境的统一视图显示器(2303)，包含一个或多个有序使用一种三维统一视图生成引擎(2306)的三维模型；

A unified view display (2303) for displaying a 3D graphics environment, consisting of one or more ordered 3D models using a 3D unified view generation engine (2306);

一个三维模型显示器(2305)，用于显示使用三维对象生成引擎(2307)模拟的对象的三维模型，三维模型显示器至少包含一个用于显示虚拟交互三维模型的显示空间；

A 3D model display (2305) for displaying a 3D model of an object simulated using a 3D object generation engine (2307), comprising at least one display space for displaying a virtual interactive 3D model;

一个虚拟操作子系统(2308)，用于提供显示三维模型的操作功能，该虚拟操作子系统根据显示对象的特点、状态和性质在加载该三维模型时安装，该虚拟子系统直接与三维模型显示器和三维对象生成引擎连接；

A virtual operation subsystem (2308) is used to provide the operation function of displaying a 3D model. The virtual operation subsystem is installed when loading the 3D model according to the characteristics, state and nature of the display object. The virtual subsystem is directly connected with the 3D model display and the 3D object generation engine;

可选择使用虚拟辅助子系统(2304)作为一种双向通讯的输入模式，该虚拟辅助子系统包括另一个图形用户界面和一个自然语言处理组件，该自然语言处理组件与中央数据库(2309)集成，用于处理用户以语词形式输入的内容，并按照收到的输入内容提供输出；

Optionally, a virtual assistance subsystem (2304) is used as an input mode for two-way communication, consisting of another graphical user interface and a natural language processing component integrated with a central database (2309) for processing user input in the form of words and providing outputs based on the input received;

可选择使用一个直播显示器(2312)，用于显示含有多个对象之处的实况，在此情况下，每个识别对象都需建立动态链接，通过该动态链接调用三维模型显示器，显示该识别对象的三维模型；

Optionally, a live display (2312) can be used to display the reality of a place containing multiple objects, in this case, each identified object needs to establish a dynamic link, through the dynamic link to call the 3D model display, display the 3D model of the identified object;

可选择使用一个摄像机(2311)，捕捉基于背景映射的交互视频，摄像机捕捉的视频放于三维模型显示器的下方；

Optionally, a camera (2311) can be used to capture interactive video based on background mapping, and the video captured by the camera is placed below the 3D model display;

本申请提供了一种数据库中的随机抽样方法，该方法包括以下步骤：将多个记录分为多个组；

This application provides a method of random sampling in a database that includes the following steps: grouping multiple records into multiple groups;

随机抽取该多个组中的一个或多个组；

Randomly select one or more of the groups;

以及在所随机抽取的该一个或多个组中，随机抽取要抽取数量的记录。

And randomly extract the number of records to be extracted from one or more randomly extracted groups.

此外，本申请还提供了一种用于数据库中的随机抽样系统和计算机可读存储介质。

In addition, the application provides a random sampling system and a computer readable storage medium for use in a database.

对所随机抽取的所述一个或多个组中的数据进行逐一扫描；

Scanning the data in one or more randomly selected groups one by one;

对于每一个所扫描的数据，基于要抽取的记录的数量，产生第二随机数；

For each scanned data, a second random number is generated based on the number of records to be extracted;

如果所述第一随机数等于所述第二随机数，则抽取所述记录。

If the first random number is equal to the second random number, the record is extracted.

基于要抽取的记录的数量产生第二随机数还基于所述多个记录的数量。

Generating a second random number based on the number of records to be extracted is also based on the number of the multiple records.

基于所述要抽取的记录的数量和所述多个记录的数量产生第二随机数是随机产生小于等于所述多个记录的数量除以要抽取的记录的数量的第二随机数。

Generating a second random number based on the number of said records to be extracted and the number of said records is randomly generating a second random number less than or equal to the number of said records divided by the number of records to be extracted.

随机抽取装置，被配置为：随机抽取所述多个组中的一个或多个组，并在所随机抽取的所述一个或多个组中，随机抽取要抽取数量的记录。

A random extraction device is configured to randomly extract one or more of the groups and randomly extract the number of records to be extracted from the randomly extracted one or more of the groups.

本发明公开了一种保护隐私文件的方法和装置。

The invention discloses a method and a device for protecting a private file.

所述方法包括：获取需要隐藏的隐私文件信息，拦截文件访问操作，当拦截到文件访问操作时，获取该文件访问操作针对的文件信息，将该文件访问操作针对的文件信息与所述隐私文件信息进行比对，判断该文件访问操作针对的文件中是否包含隐私文件，如果包含，则从该文件访问操作获取的文件中移除隐私文件，使隐私文件对该文件访问操作不可见。

The methods include:

本发明的技术方案，由于直接对文件访问操作进行拦截，从其获取的文件中移除隐私文件，因此即使通过系统设置或者其他软件的遍历也看不到隐藏文件，从而保护了用户隐私文件的安全。

Because the technical scheme of the invention directly intercepts the file access operation and removes the private file from the obtained file, the hidden file cannot be seen even through the system setting or other software traversing, thus protecting the security of the user's privacy file.

当拦截到文件访问操作时，获取该文件访问操作针对的文件信息；

When a file access operation is intercepted, information about the file targeted by the file access operation is obtained.

将该文件访问操作针对的文件信息与所述隐私文件信息进行比对，判断该文件访问操作针对的文件中是否包含隐私文件；

Compare the file information targeted by the file access operation with the privacy file information to determine whether the file targeted by the file access operation contains the privacy file;