Database Management & Data Analytics
A database is a collection of data or records. Database management systems are software systems used to store, retrieve and run queries on data. DBMS, i.e. Database Management Systems uses a standard method to store and organize data. The data can be added, updated, deleted, or traversed using various standard algorithms and queries.
Types of DBMS:
• Hierarchical databases
• Network databases
• Relational databases
• Object-oriented databases
• Graph databases
• ER model databases
• Document databases
Hierarchical Databases
The IBM Information Management System (IMS) and Windows Registry are two popular examples of hierarchical databases.
Hierarchical Database System Structure was developed by IBM in the early 1960s.
Here, data gets stored in parent-children relationship nodes. Besides the actual data, records also contain information about their groups of parent/child relationships.
As the name suggests, ‘Hierarchy’, it is a tree-like structure. There are links attached between each record as a parent-child/children relationship.
Data gets stored in the form of a collection of fields in which each field contains only one value, i.e., every individual record has only one parent and a parent can have one or more than one children. Hence, to retrieve the data, we need to traverse each tree until we get the desired data.
It is simple but inflexible due to the parent-child one-to-many relationship.
Hierarchical Databases are generally used in Banking and Telecommunications industries to build high performance and manage applications.
Advantages of Hierarchical Databases
The hierarchical database has the advantage that it is like a double-edged sword and can be retrieved and manipulated or altered rapidly due to the tree-like structure of the record and also relationships between records are defined in advance.
Disadvantages of Hierarchical Databases
The disadvantage here is that, each node can only have one parent, and children can’t have relationships or linkages between them, even if they make sense from a logical standpoint.
In order to add record or data in the database, the whole database needs to be redefined.
Network databases
The network database structure was invented by Charles Bachman. Network database management systems (Network DBMSs) uses network structure to create a relationship between entities.
Network databases are mainly used on a large network of computers.
Network databases are similar to hierarchical databases except for one key point. Unlike Hierarchical DBMS, in network databases one node can have a relationship with multiple entities.
In network databases, parents are termed as occupier and children are termed as members. Data in the network database is organized as many-to-many relationships.
Integrated Data Store (IDS), IDMS (Integrated Database Management System), Raima Database Manager, TurboIMAGE, and Univac DMS-1100 are a few popular examples of Network DBMS.
Relational databases
Relational Databases are the most popular among all databases.
In this type of database, the data is relational and is stored in a tabular from in rows and columns. Each row represents record and column represents the attribute.
Every individual field represents the data value. In order to query the Relational
Databases, Structured Query Language (SQL) is used which includes insertion, deletion, manipulation and search of the records.
Relational database depicts the relation between two or more tables, through Key Fields. Every row has its unique key field and these key fields are used to connect one table to another one.
Some of the popular Relational DBMS users are Oracle, SQL Server, MySQL, SQLite, and IBM DB2.
There are two major advantages of Relational DBMS:
Relational databases can be used with little or no training.
Database entries can be modified without specifying the entire body.
Properties of Relational Tables
Values are Atomic
Each row is individual.
Column Values are the same thing.
Columns are undistinguished.
Sequence of Rows is Insignificant.
Each Column has a common name.
Object-Oriented databases
Object-oriented Databases were created in the early 1980s. It deals with the functionality of the object-oriented programming and increases the semantics of the C++ and Java.
Advance programming language objects are required in Object-Oriented Databases.It provides full-featured database programming capability while containing native language compatibility such that it alters the database functionality to object programming languages.
Applications in Object-Oriented databases require less code, use more natural data modeling, and code bases are easier to maintain.
Object developers can write complete database applications with a less amount of effort and in less time.
Object-oriented databases use small software called objects. The objects themselves are stored in the object-oriented database.
Every object in Object-Oriented databases contains two elements:
Piece of data (e.g., sound, video, text, or graphics)
Instructions or software programs called methods
Some Object-Oriented Databases were designed to work with OOP languages such as Delphi, Ruby, C++, Java, and Python. Some popular Object-Oriented Databases are TORNADO, Gemstone, InterSystems Cache, Versant Object Database, ODABA, ZODB, Poet. JADE, and Informix. One of the most important advantages of Object-Oriented Database is its ability to mix and match reusable objects. It provides the unbelievable multimedia capability. However, the disadvantages are that They are more expensive to develop. Most organizations don’t want to abandon their traditional database approach and convert into this database system.
Graph databases
Graph databases are NoSQL databases and it uses the graphical structure for semantic queries.
Data is stored in the form of nodes, edges and properties in which a node is equivalent to a record, the edge is a link between two nodes and properties are additional information added into the nodes.
Neo4j, Azure Cosmos DB, SAP HANA, Sparksee, Oracle Spatial and Graph, OrientDB, ArrangoDB and MarkLogic use Graph DBMS
Graph database are supported by some RDBMs, including Oracle and SQL Server 2017 and later versions.
ER model databases
Entity-Relations Model Database was developed by Peter Chen 1976.
Here, an ER model is typically implemented as a database. In a simple relational database implementation, each row of a table represents one instance of an entity type, and each field in a table represents an attribute type. In a relational database, a relationship between entities is implemented by storing the primary key of one entity as a pointer or “foreign key” in the table of another entity.
Document databases
Document databases (Document DB) are also NoSQL databases that store data in the form of documents. Each document represents the data, its relationship between other data elements, and attributes of data. Document database store data in a key-value form.
Document DB has become popular recently due to their document storage and NoSQL properties. NoSQL data storage provides a faster mechanism to store and search documents.
Popular NoSQL databases are Hadoop/Hbase, Cassandra, Hypertable, MapR, Hortonworks, Cloudera, Amazon SimpleDB, Apache Flink, IBM Informix, Elastic, MongoDB, and Azure DocumentDB.
Data analysis is a process of inspecting, cleansing, transforming and modelling data with the goal of discovering useful information, informing conclusions, and supporting decision-making.
The kinds of insights you get from your data depends on the type of analysis you perform. In data analytics and data science, there are four main types of analysis: Descriptive, diagnostic, predictive, and prescriptive.
To simplify it further…
Descriptive: ‘what is happening in your business?’
- Comprehensive, accurate and live data
- Effective visualisation
This is the most common of all forms. In business it provides the analyst a view of key metrics and measures within the business.
A monthly profit and loss statement is a good example of descriptive analysis. Utilising effective visualisation tools enhances the message of descriptive analytics.
Diagnostic: ‘Why is it happening?’
- Ability to drill down to the root cause
- Ability to isolate all confounding information
Descriptive analytics, the initial step in most companies’ data analysis, is a simpler process that chronicles the facts of what has already happened. Diagnostic analytics takes it a step further to uncover the reasoning behind certain results.
Diagnostic analytical tools will empower an analyst to drill down and in so doing isolate the root-cause of a problem.
Predictive: ’What’s likely to happen?’
- Business strategies have remained fairly consistent over time
- Historical patterns being used to predict specific outcomes using algorithms
- Decisions are automated using algorithms and technology
The term predictive analytics refers to the use of statistics and modeling techniques to make predictions about future outcomes and performance. Predictive analytics looks at current and historical data patterns to determine if those patterns are likely to emerge again. This allows businesses and investors to adjust where they use their resources to take advantage of possible future events.
Prescriptive: ‘What do you need to do?’
- Recommended actions and strategies based on champion/challenger testing strategy outcomes
- Applying advanced analytical techniques to make specific recommendations
In essence, prescriptive analytics takes the “what we know” (data), comprehensively understands that data to predict what could happen, and suggests the best steps forward based on informed simulations.