Data Modelling [eur ( LEARNING OBJECTIVES Introduction at is Data Modelling? ._Types < of Data Models Data Modelling Techniques INTRODUCTION We have now gone over the various concepts of business analytics. In previous chapters, we went over the data warehouse, data mining, and data visualisation depth. Let’s talk about data modelling now. Before we get into the specifics of data modelling, let’s define modelling/Modelling is any representation of omething else, usually on a smaller scale. To plan offensives, military verything, whether it’s inflation, marketing strategy, or consumer purchasing behaviour toward a product or the stock market, changes over time. As a ult, the data changes to reflect how the world works. Data enables us to arch illnesses and their treatments, as well as business revenue growth pplication in certain way. S| useless | to the production Separeneng:This is where the use of Data Modelling comes into play. It applies relations) juidelines to data? A Data Model simplifies data but Tistead ene tun: SSlaable atormatign which companies can uilize for decision-making ang Strategic planning. In this chapter, we will learn about data modelling in depth, 9X WHAT When data is collected, it is in its most basic form, As data analysts, we must extract Useful information from data and analyse itusing proper tools. Before embarking on Tata modelling, businesses should ensure that data they” ig mects basic : accuracy and interpret therwise, the data Tesult, before embarki [A Data Model is an abs Conceptual models es Given the foregoing, it is the first and most important step in defining the structure of available data. Every company has its own set of requirements. ‘Ks a result, cach business will have a unique set of data. Data is needed based ‘on feedback from stakeholders so that it can be implemented into the development of a new framework or adapted in a version of an existing one. These business requirements are then converted into data models, which are sed fo create a concrete database design. Data modelling enhances data alytics. Data modelling is a process which will provide us with a graphical Tepresentation of a various steps that must be completed, beginning with data Collection and ending with results, Data modelling can be of the entire information or a subset of it. The goal of data modelling is to demonstrate the relationship between various steps to be taken, the various ways data can be grouped and organised, as well as its formats and attributes. It entails communicating data and information using both symbols and language. The «data modelling process is an effort to comprehend and define the data that the business requires. oe TPES OF DATA MODELS Database design, like any other process, begins with a high degree of abstraction and progresses to become more specific and precise. In this contex’ DATA MODELLING? mei ‘displaying” only the relevant attributes of information ‘ata and “hiding” the irrelevant details, () Bede Modelling Inetne 12 Communicate bepwen Busines PROP, wrinn ros Khe vera bys Omd tre! Lech Whe! Con AWLALL Welt al peop VOUS 4Ppeeweeeins, EX DOMPLE OF RBCTEME HON) z= 93 wwe use Microsoft Word or Excel, for example, we are only concerned the tools available in these programmes. We're not concerned with how tools, such as how to cut and paste text, are created. We're simply not faterested in the details Aynat concerns us isthe “abstract” view of these applications that will assist ve in completing our task. This is a simple example of abstraction. Thus, Microsoft Word or Excel contain all ofthe tools, but only from our viewpoint, Timply beeause we are only interested in the abstract view ofthese applications. ication user does not need to be concerned about how a particular feature as_neede abstraction, we mostly create complex data types and hide their ‘aiplementation, We only expose the operations for manipalating these data “ without going into the implementation details. Rey BE a Be MoREt iy ‘Data models are generally classified into three types bastd on their level of complexity. The procedure will begin with a conceptual model, then move on to logical model, and finally to a physical model, DATAMODELING ls of Data Modeling inane being no Leape OF 0 Model, ay. : 93.1 Conceptual Data Model roy 20 ALEVtbe! “Anu Semantler © A conceptual data model defines the el, or overall business and ‘ da gs It depicts Srzanteational concepts, sich as what the system COTO Sntains, how it will be organised, and which business rules are involved. The conceptual model is created as a component of the initial project Tequirement phase process. It is more concemed with establishing entities, asva business Analytics well as demonstrating their relationship and characteristics, rather than with database details For example, every business maintains a database of its customers, employees, and products. Entities in business are customer, employee, or product databases. These entities ae a Entities and their relationships are, defined in your conceptual model. ; * Jvtrus model, truly ow ac eens ob tats columns, Obici OMLenLid Logie, as the name suizgests, is a Serjes of statements/each of which must be homes, oaitihe mnt before itis true] “It it rains today, roads will be slippery.” ME G&S, for example. It is a series of statements that if jt rains today, then only the clocumnveyroads will be slippery; otherwise, they will nat) Similarly, the logical data a Jyusmodel is built on a series of statements. The first statement is the previous AWCOVC mentioned conceptual model. A logical data model extends the conceptual model by including specific data characteristics within each entity and specific relationships between those attributes. It goes into greater detail about the concepts, Investor A, for example, purchases shares B from portfolio investor C. This isa technical model of the rules and data structures as defined by data architects and business analysts, and it will help you decide what physical model your data and business requirements require. The logical data model's goal is to create a technical map of rules and data structures. Logical data models can be useful in highly procedural implementation environments or for data-driven projects like dala warchouse. design oF tepertine soot warehouse, design_or reporting system \ development. 7 Unu moked cones about the ' 9.3.3. Physical Data Modelg “AOU Pry weal crructure 10 { tre partion CRU, — They define the structure of the data that will be physical stored hares a wcts, They provide a fini ign that can be i ied_as_a relational . . database, including associative tables she relationships between (todo ebitities as well as the primary ke those_relationships alive. Physi As saysical monn , management system (DBMS)-spx