Watson Analytics 2.0

Continuing from where we left off the last time. We covered two out of the five section of the software indicated in the diagram below by the bits in grey. So we are discussing 'Assemble' to 'Refine'.

Assemble: With 'Assemble' making a dashboard has never been easier. The software has reduced making of charts to a  simple 'drag and drop' process. The section has templates and layouts that users can use to make all kinds of dashboards. Once the template and layout is selected, fields are simply dragged into axes and filters to make graphs as shown in the diagram below.

There are several options of charts that can be made, it is just as simple as clicking on the chart that meets the objective and the job is done. Below is a diagram showing the types of charts available.

Social Media: This section scrapes social media for data based on criteria specified by the user. The criteria will include fields like language, source,The data collected can analyzed by topics,theme, geography,sentiment, active authors and demographics. This tool is useful for finding out the buzz around a service (or product or may be even a topic) in social media.  The software gives clues from data generated around the chosen topic in form of a word cloud, just like in the diagram.

Users can include these clues as  terms closely related to the search topic to get better results, they can also break down the topic by specifying a criteria of some sort, this is known as a 'Theme' in Watson Analytics. After the search is all sorted, a dataset is generated which can now be analyzed.

The diagram above shows a 'Sentiment' chart from the data collected.  The dataset fields are displayed at the bottom of the chart. Other information like topics, themes, languages, dates, documents and mentions are displayed at the top. All these details will specify the search and scrape criteria for the data. Users can change charts by using the chart criteria to display charts for criteria that was earlier specified.

Refine: Watson Analytics assesses data before use, result of assessment are usually displayed on the tile that links to the data as shown in the diagram below.

 It assesses data for missing values and inconsistencies that may affect use of the data and then awards marks. 100 for data that is consistent and has no missing values. The 'Refine' section does exactly what it says on the box. It is used for cleaning, munging and editing of data. The section can enable users to group data, form hierarchy, do calculations. Users can also search data, include/exclude rows and columns.  Each column is assessed , marks and charts are displayed showing quality of data.

There is still much to learn in Watson Analytics, the blog is just a headstart. As always, there's more information on the internet.

Give Watson Analytics a try. Thanks for reading.