Data Blending

Data Blending and ETL

This workflow illustrates how to blend different data sources within KNIME Analytics Platform. The resulting blended table is the source of the Bar Chart node used to visualize the average age of the customers for each product.

Find more information on KNIME’s Learning page at http://www.knime.com/learning (courses, tutorials, cheatsheets, books, and more)

Task

Blend together various data sources in an effective data integration process using ETL (Extract, Transform and Load). Visualize the final result.

Data Reading

6 data sources:

- CSV file - Webdata Old System.csv (Old Web System)

- SQLite Database - WebActivity.sqlite (New Web System)

- .table file - Sentiment Analysis.table

- CSV file - Sentiment Rating.csv

- CSV file - Demographics.csv

- Excel file - Prooduct Data2.xls

The files are located in TheData/Misc/

Data Blending & ETL

- Concatenate web activity data from old and new systems

- Replace sentiment labels (text) with numeric scores

- Set all product names to lowercase

- Join all data together by the customerKey column

- Group by products, and aggregate by age (mean)

Visualize

Create a bar chart showing the average age of customers for each product