First, we cover creating an infrastructure or an object model that starts with a request, access definition, and column maps for data extraction.
首先，我们创建一个基础或者对象模型，从分析需求、访问定义，以及数据提取的列映射开始。
Specially, the paper presented the creating of dimension table, fact table of multi-dimension data model, analyzed the construction of data model, data extraction and data maintenance tools.
其中重点介绍了多维数据模型的维表、事实表的结构设计，分析了数据模型的构建、数据抽取工具和数据维护工具的设计及实现。
This model including three modules: the data pretreatment, the attribute reduction and the rule extraction, then confirms this model's feasibility using the example.
该模型包括数据预处理、属性约简和规则提取三个模块，并利用算例验证该模型的可行性。
In addition, the data stream parsing, extraction protocol characteristics, the establishment of ATM, IP protocol type algorithm model for rapid identification.
另外，对数据流进行解析，提取协议特征，建立A TM、IP协议类型快速识别算法模型。
This paper proposes a new algorithm using hidden Markov model for information extraction based on multiple templates due to the variety of training data.
针对训练数据来源的多样化，提出了基于多模板隐马尔可夫模型的文本信息抽取算法。
Facial features extraction algorithms presented are based on deformable template model. In this paper, facial features are described accurately about contour but not express simply with data.
根据动态模板匹配理论，提出了提取人脸面部特征的系列算法，特征是精确的轮廓描述而非简单的数字表达。
In this paper, a basic definition of the data extraction process has been given and Described a page generation model of the data extraction.
该文给出了数据抽取过程中需要的基本定义，描述了数据抽取所基于的页面生成模型。
It involved the creation of the data warehouse, the extraction of data, the achievement of OLAP, and in particular it explains the model base and method base in the system.
主要涉及到数据仓库的创建、数据析取，OLAP分析的实现。并对系统中模型库和方法库进行了详细说明。
This dissertation selected a pavilion, reconstructed its model with scanning data, and completed to draw the maps using the feature lines and contour obtained with semiautomatic extraction.
本文以凉亭为扫描对象，对扫描得到的数据进行模型重建，并利用半自动提取出的特征线与轮廓线，完成该建筑各图件的绘制。
Then introduced the transformation process that from the source code to the abstract data model. Focus on the code extraction patterns and the analysis of the system source code process.
之后介绍了源代码向抽象数据模型的转化过程，重点阐述了代码抽取模式和系统源代码分析过程；
After class definition, feature extraction, data annotation, model training and experimenting, the output proves that acceptable performance of answer extraction could be reached.
通过类型定义、特征表示、数据标注、模型训练、实验验证等一系列过程，最终的结果表明能够对论坛数据实现高性能的答案抽取。
After class definition, feature extraction, data annotation, model training and experimenting, the output proves that acceptable performance of answer extraction could be reached.
通过类型定义、特征表示、数据标注、模型训练、实验验证等一系列过程，最终的结果表明能够对论坛数据实现高性能的答案抽取。
应用推荐