Data Access and Data Wrangling with TIBCO Spotfire®
Last updated:
12:05am Feb 10, 2017

 

  Overview

This is your guide to getting your data it into a form convenient for visualization or analysis.  That includes actions like accessing, cleaning, joining, limiting, summarizing and transforming data.   

Data Access

Data can be accessed in Spotfire via the clipboard, files, information links, data connectors, data functions and other sources such as databases.  The self-service data connectors allow Spotfire users to easily connect to and analyze data from relational databases, cubes, OData sources, Hadoop and other big data sources.

Data Wrangling

  • Read the Data Wrangling data sheet - Successful analytics depends on structured, accurate, and well-formatted data. Most analytic projects require that up to 80% of their time be spent wrangling data for the purposes of analysis because data is messy, contains errors, or is not organized for analysis. Spotfire® uniquely offers inline data wrangling through a unified visual analysis and preparation interface to help you rapidly clean data while performing analysis.
  • Data handling Quick Reference Topics: Choose the Data Handling section for information on Linked to Source option for data tables, Replacing data tables and pivoting data
  • Limiting Data Quick Reference Topics:  Choose the Limit Data section for information on Filtering and Details visualizations drill-downs
  • How To Combine (Join) Data From CSV Files In Spotfire - video
  • Data Wrangling capabilities in Spotfire 7.6 - video:  An important part of visual analytics is being able to wrangle data into the shape and quality that you need.  Using data from the Google Trends Datastore, this video shows how you can do commonly needed data prep actions like removing bad records, unpivoting datasets, and splitting columns into different dimensions, all quickly and easily.
  • Visual overview of data table structures (page 18): It is sometimes challenging to understand which data sources and what methods have been used to create combined data tables. To solve this problem, data table data sources and operations can now easily be viewed in the Source view of the expanded data panel. It is possible to see detailed information about operations and preview intermediate resulting data tables after individual steps
  • Split columns into new columns based on column values:  Sometimes, column values contain multiple pieces of information. Examples are first and last name, or city and zip code. It's now easy to split columns of this type into separate columns containing the individual values from the original column. The original column can then be hidden from the analysis, not to distract and take up valuable space (in, for example, the Data panel).
  • Unpivot from the data panel:  Data can be organized in different ways, for example, in a short/wide or tall/skinny format, but still contain the same information. Often, it is easier to visualize data organized in a tall/skinny format, that is, when the values are collected in just a few value columns. Unpivoting is one way to transform data from a short/wide to a tall/skinny format, so the data can be presented the way you want it in the visualizations. The Data panel (both in TIBCO Spotfire Analyst and TIBCO Spotfire Business Author) now has a built-in unpivot tool on the right-click menu.
  • Using the OVER function
    • Here is a great video on using the OVER function in calculated columns
    • Here is an article showing use of OVER in Custom Expressions:

Content Analytics

Spotfire partners with Attivio to provide text analytics for unstructured content from emails, documents, notes, support tickets, weblogs and social media feeds in 28 different languages.

 

 Back to Main Spotfire Wiki page