数据科学和数学建模

资料建模 (Data Modeling)

Data modeling is the method of documenting a fancy computer code style as simply understood diagram, victimization text, and symbols to represent the method knowledge has to flow. The diagram may be accustomed to guaranteeing economical use of knowledge, as a blueprint for the development of the latest code or for re-engineering an inheritance application.

数据建模是一种记录精美的计算机代码样式的方法，该代码以简单易懂的图表，受害文本和符号的形式表示，这些知识代表了知识的流动。 该图可能习惯于保证知识的经济使用，作为开发最新代码或重新设计继承应用程序的蓝图。

Data modeling is a crucial ability for data scientists or others involved in data analysis. Data models are designed throughout the analysis and style phases of a project to confirm that the need for a brand new application is understood. Data models also can be invoked later within the data lifecycle to rationalize data styles that were created by programmers on a commercial ad-hoc basis.

对于数据科学家或其他参与数据分析的人员而言，数据建模是一项至关重要的能力。 在项目的整个分析和样式阶段都设计数据模型，以确认已了解对全新应用程序的需求。 数据模型也可以稍后在数据生命周期内调用，以合理化程序员在商业临时基础上创建的数据样式。

数据建模方法 (Data modeling approaches)

Data modeling may be a conscientious direct method and, as such, is usually seen as being at odds with fast development methodologies. As Agile programming has acquired wider use to hurry development comes, after-the-fact strategies of data modeling are being tailored in some instances. Typically, a data model may be thought of as a flow diagram that illustrates the relationships among data. It permits stakeholders to spot errors and build changes before any programming code has been written. or else, models may be introduced as a part of reverse engineering efforts that extract models from existing systems, as seen with NoSQL knowledge.

数据建模可能是一种认真的直接方法，因此，通常认为它与快速开发方法不符。 随着敏捷编程越来越广泛地用于加快开发速度，在某些情况下，正在量身定制数据建模的事后策略。 通常，数据模型可以被认为是说明数据之间关系的流程图。 它允许利益相关者在编写任何编程代码之前发现错误并进行更改。 否则，可以将模型作为逆向工程工作的一部分引入，如从NoSQL知识中看到的那样，逆向工程从现有系统中提取模型。

Data modelers usually use multiple models to look at an equivalent knowledge and make sure that all processes, entities, relationships, and data flows are known. They initiate new comes by gathering needs from business stakeholders. Data modeling stages roughly break down into the creation of logical data models that show specific attributes, entities, and relationships among entities and therefore the physical data model.

数据建模人员通常使用多个模型来查看同等知识，并确保所有流程，实体，关系和数据流都是已知的。 他们通过收集业务利益相关者的需求来发起新的挑战。 数据建模阶段大致分为创建逻辑数据模型，这些逻辑数据模型显示特定的属性，实体以及实体之间的关系，从而显示物理数据模型。

The logical data model is the premise for the creation of a physical data model, that is particular to the appliance and information to be enforced. A data model will become the premise for building a lot of elaborate data schema.

逻辑数据模型是创建物理数据模型的前提，该物理数据模型特定于要实施的设备和信息。 数据模型将成为构建大量复杂数据模式的前提。

分层知识建模 (Hierarchical knowledge modeling)

Data modeling as a discipline began to arise within the Sixties, concomitant the upswing in the use of direction systems (DBMSes). Data modeling enabled organizations to bring consistency, repeatability and regular development to processing. Application finish users and programmers were ready to use the information model as a reference in communications with data designers.

六十年代， 数据建模作为一门学科开始兴起，随之而来的是定向系统(DBMS)的使用。 数据建模使组织能够为处理带来一致性，可重复性和常规开发。 应用程序完成用户和程序员已准备好将信息模型用作与数据设计人员进行通信时的参考。

Hierarchical knowledge models that array data in arboreal, one-to-many arrangements marked these early efforts and replaced file-based systems in several fashionable use cases. IBM's Data Management System (IMS) may be a primary example of the ranked approach, that found wide use in businesses, particularly in banking. Though ranked data models were for the most part outdated -- starting within the Eighties -- by relative data models, the ranked methodology is common still in XML (Extensible Markup Language) and geographic data systems (GISes) these days. Network data models additionally arose within the period of DBMSes as a way to produce data designers with a broad abstract read of their systems. One such example is that the Conference on knowledge Systems Languages (CODASYL), which shaped within the late Fifties to guide the event of a typical programing language that would be used across varied styles of computers.

以树状，一对多排列的方式排列数据的分层知识模型标记了这些早期工作，并在几个流行的用例中替换了基于文件的系统。 IBM的数据管理系统(IMS)可能是排名方法的主要示例，该方法在企业(尤其是银行业务)中得到了广泛的应用。 尽管从80年代开始，相对数据模型在大多数情况下已过时(相对于80年代而言)，但是如今，这种排序方法仍然在XML(可扩展标记语言)和地理数据系统(GIS)中很常见。 在DBMS时代，网络数据模型也应运而生，这是一种使数据设计人员对其系统有广泛的抽象理解的方法。 一个这样的例子就是知识系统语言会议(CODASYL)，该会议在五十年代后期形成，旨在指导一种典型的编程语言的使用，该编程语言将在各种样式的计算机上使用。

关系知识建模 (Relational knowledge modeling)

While it reduced program quality versus file-based systems, the ranked model still needed an elaborate understanding of the particular physical knowledge storage used. Planned as another to the ranked data model, the relative data model doesn't need developers to outline data ways. Relative data modeling was 1st represented in an exceedingly 1970 technical paper by IBM research worker E.F. Codd. Codd's relative model set the stage for business use of relational databases within which data segments are expressly joined by use of tables, as compared to the ranked model wherever data is implicitly joined along. Presently when its origination, the relative data model was plus the Structured source language (SQL) and commenced to realize an ever-larger foothold in enterprise computing as an economical suggests that to method data.

尽管与基于文件的系统相比，它降低了程序质量，但排名模型仍然需要对所使用的特定物理知识存储有详尽的了解。 作为排名数据模型的另一个计划，相对数据模型不需要开发人员来概述数据方式。 相对数据建模在IBM研究工作者EF Codd于1970年发表的一篇技术论文中排名第一。 Codd的相对模型为关系数据库的业务使用奠定了基础，在该数据库中，通过隐式联接数据的排名模型与通过表的使用将数据段明确联接在一起。 目前，当相对数据模型开始使用时，它就加上了结构化源语言(SQL)，并开始在企业计算中实现越来越大的立足点，因为经济地建议对数据进行处理。

实体关系模型 (The entity-relationship model)

Relational data modeling took another revolution starting within the mid-1970s as the use of entity-relationship (ER) models became a lot of prevailing. Closely integrated with relative data models, ER models use diagrams to diagrammatically depict the weather in an exceedingly information and to ease understanding of underlying models.

关系数据建模在1970年代中期开始发生了另一次革命，因为实体关系(ER)模型的使用变得非常流行。 ER模型与相关数据模型紧密集成，使用图表以过度信息的形式示意性地描述天气并简化对基础模型的理解。

With relative modeling, data varieties are determined and infrequently modified over time. Entities comprise attributes; as an example, An entity's attributes might embody the name, first name, years used than on. Relationships are visually mapped, providing a prepared suggests that to speak knowledge style objectives to varied participants in data development and maintenance. Over time, modeling tools, as well as Idera's ER/Studio, Erwin data creator and SAP PowerDesigner, gained wide use among knowledge architects for planning systems.

通过相对建模，可以确定数据种类，并且不经常修改数据种类。 实体包括属性； 举例来说，实体的属性可能包含名称，名字，使用年限。 关系被可视化地映射 ，提供了一个准备好的建议，可以向数据开发和维护中的各个参与者说出知识风格的目标。 随着时间的流逝，建模工具以及Idera的ER / Studio，Erwin数据创建者和SAP PowerDesigner在计划系统的知识架构师中得到了广泛的使用。

As object-oriented programming gained ground within the Nineties, object-oriented modeling gained traction thus far in our way to style systems. Whereas bearing some alikeness to ER strategies, object-oriented approaches dissent therein they specialize in object abstractions of real-world entities. Objects are sorted in school hierarchies, and therefore the objects among such category hierarchies will inherit attributes and strategies from parent categories. As a result of this inheritance attribute, object-oriented data models have some benefits versus ER modeling, in terms of guaranteeing data integrity and supporting a lot of complicated data relationships. Additionally, arising within the Nineties were data models specifically orienting toward data reposition wants. Notable examples are snowflake schema and star schema dimensional models.

九十年代，随着面向对象编程的发展，迄今为止，面向对象的建模在我们设计系统的方式中获得了吸引力。 尽管与ER策略有些相似，但在其中却没有采用面向对象的方法，它们专门研究现实世界实体的对象抽象。 对象按学校层次结构排序，因此，此类类别层次结构中的对象将从父类别继承属性和策略。 由于具有这种继承属性，因此在保证数据完整性和支持许多复杂的数据关系方面，面向对象的数据模型相对于ER建模具有一些优势。 此外，在90年代出现了专门针对数据重新定位需求的数据模型。 值得注意的示例是雪花模式和星形模式尺寸模型。

图形数据模型 (Graph data models)

A branch of ranked and network data modeling is that the property graph model, which, at the side of graph databases, has found hyperbolic use for describing complicated relationships among knowledge sets, notably in social media, recommender and fraud detection applications.

排名和网络数据建模的一个分支是属性图模型，该属性图模型在图数据库的一侧发现了双曲线的用法，用于描述知识集之间的复杂关系，特别是在社交媒体，推荐器和欺诈检测应用程序中。

Using the graph knowledge model, designers describe their system as a connected graph of nodes and relationships, very much like they may do with ER or object data modeling. Graph data models may be used for text analysis, making models that uncover relationships among knowledge points among documents.

使用图知识模型，设计人员将他们的系统描述为节点和关系的连接图，非常类似于他们对ER或对象数据建模所做的工作。 图形数据模型可用于文本分析，使模型能够揭示文档之间知识点之间的关系。

翻译自: https://www.includehelp.com/data-science/data-modeling.aspx

数据科学和数学建模