Sunday 11 October 2015

Data Visualization

Information in an unorganized form is called data. Data can come from diverse sources such as social media, sensors, transaction logs etc. Tables or text files are used to store the data. But it is not possible to understand the data in these different formats. It is proven that human brain understands visuals rather than facts and figures. The data visualization tools are used to understand the data. Wide variety of visualization tools are available in the market. Some of them include Actuate, QlikView, Spotfire, Google Chart API, Flot, Raphael etc. Besides the common functionalities each tool provides its own features. So the choice of a tool depends on the context in which it is used. Data consist of raw facts and figures. Data visualization is the process of representing the data in the form of charts, maps or any other graphical means which makes the content easier to understand. The first data visualization was created by Rene Descartes using the Cartesian co-ordinate system in the 17th century. With the advent of the Social media sites like Facebook and Twitter the amount of data collected, stored and analyzed have increased significantly. Hence data analysis has gained importance. Friend maps and Twitter Vision are the data visualizations familiar to such users. Graphical representations are more helpful than the Excel files or tables containing the data since they help to think about the data by revealing the underlying patterns and connections between different elements. Data Visualization tools enable users to quickly create complex visualizations using data from diverse data sources. Some of the leading data visualization tools include Tableau, QlikView and Actuate.

Functions of Data Visualization Tools

The following are the main functions of data visualization tools:

  • Minimization of effort: The data can be analyzed quickly by connecting it to different sources using the Drag and Drop   functionality. This reduces the Lines of Code written by the developers.
  • Framing Questions: Data visualizations help in identifying outliers in the data. This leads to identifying the problems in the data.
  • Answering Questions: The findings in the visualization can be used identify the trends. These can be used to predict the future observations.

Requirements for data visualization

The requirements of Data visualization can be classified into following:

  • Functional Requirements


Functional requirements includes the set of activities that the system should do. It includes the following:
  1. Support real time creation of dynamic and interactive charts
  2. Allow interaction of multiple users with the data across diverse platforms
  3. Ability to visualize data from different data sources
  4. Provide secure access of data by end users


  • Non Functional Requirements


Non functional requirements are those requirements that are not part of the functional requirements. They are mainly used to judge the performance of the system. The set of non functional requirements consist of Performance, Scalability, data Integrity etc.

Stages of Data Visualization

Benjamin Fry is an American expert in Data visualization. He has proposed seven stages in data visualization. Each of the stages can be briefly explained as follows:

  1. AcquireThe data must be retrieved from a data source.
  2. ParseIt is not necessary that the data obtained will be in a format suitable for visualization. Hence the data must be structured into categories.
  3. FilterThe unimportant data must be removed to prevent information overload.
  4. MineDifferent statistical methods can be applied to identify the trends and patterns in the data.
  5. RepresentDifferent views and representations of the data leads to better decision making.
  6. RefineThe basic visual model chosen will be further refined to make the representation clearer and visually intuitive.
  7. InteractAdd different methods of interaction to allow users to decide what they see and how they see.

Data Visualization Tools

Some of the leading data visualization tools are the following:

Actuate

The Actuate Data Visualization Suite consist of the BIRT Analytics, BIRT Designer and BIRT iHub Runtime and Viewer.

  • BIRT Analytics: It is a visual data mining and predictive analytics tool. The main features include:
Social- It can connect to both social and web data sources including Facebook, Twitter              and Google Analytics.
Predictive- It incorporates both predictive analysis and visual data mining in a single                    product.
Quick Big Data- It can analyze huge volume of data within short span of time.
  • BIRT Designer: It is an open source reporting software based on the Eclipse IDE. BIRT Designer is used by developers to create visualizations based on the data from different data sources. It has the following characteristics:
          - Data Integration from diverse data sources
          - Consist of tools to secure, filter, format and present dynamic reports to end user
          - BIRT Designer includes a set of component libraries
  • BIRT iHub Runtime and Viewer: It is the deployment platform for all the BIRT content. It includes the following functionalities:
        - It consist of data drivers to data sources such as Hadoop and Oracle
        - Publishes BIRT content to web, mobile and other print media
        - Controls the access to the BIRT content

QlikView

QlikView is a software that helps users to retrieve and analyze data easily from any source. It offers wide variety of charts, tables etc. for representing the data. The QlikView stack of products include QlikView Personal Edition, QlikView Server and QlikView Publisher.

  • QlikView Personal Edition: It provides full QlikView functionality, but it is not possible to open documents created by other users. To do this we need a QlikView license. QlikView Personal edition can be downloaded as a standalone application.
  • QlikView Server: QlikView information can be shared and hosted using the QlikView Server platform.
  • QlikView Publisher: It manages the content and access. QlikView Publisher distributes data stored in QlikView documents to end users.

Tableau

Tableau helps to drag and drop data to visualize it. It consist of Tableau Desktop, Tableau Server, Tableau Online, Tableau Public and Tableau Reader.

  • Tableau Desktop is a standalone desktop application
  • Tableau Server is a browser based business intelligence solution.
  • Tableau Online is a hosted version of tableau server.
  • Tableau Public is a service used for interactive data visualization.
  • Tableau Reader is a free desktop application used to view the visualizations built in tableau desktop.

With the advent of Social media sites and search engines, large amount of data is produced daily. The urge for data analysis is increasing. So it is high time to analyze raw data and present the information to the end user in an intuitive way. Besides the wide variety of tools available, the evaluation of the nonfunctional requirements is done for Actuate, QlikView and Tableau. The most important feature of the Actuate is its Live Excel functionality which helps the data to be exported as pivot tables. QlikView has an intuitive user interface but the implementation time is high compared to Tableau. Clearly each tool has its own USP and many of their NFR attributes complement each other.


Choose a job you love, and you will never have to work a day in your life.