The Reporting 2013 Conference is coming later this month in Orlando, so I thought I would use Text Analysis from Data Services to see what is being discussed at the conference – themes, products, organizations.  I use the output from Text Analysis and take that to Lumira.

First I took the brochure and copied the text to Notepad so it was all raw text.  Please note I did NOT copy the entire brochure but just a few pages.  This is how it looks below:

/wp-content/uploads/2013/11/1afig_313147.png

Clearly the above shows it is unstructured text.  How to analyze?

Then I went to Data Services and used the Base Entity extract transform and read the file in:

/wp-content/uploads/2013/11/1fig_313157.png

In the above it is taking the text file in, using the Base Entity extract transform to determine what to read, and then it will output it to a Text file.

/wp-content/uploads/2013/11/3fig_313158.png

In Data Services I am using entity types NOUN_GROUP, ORGANIZATION, and PRODUCT

3inlumira.PNG

I take the file to Lumira and read it in.

4wordcloudproductmentions.PNG

I decide to look at a word cloud with product mentions; predictably HANA has top mention and then Web Intelligence and Predictive Analysis.

5top10organizations.PNG

I look at the word cloud with organization mentions and SAP is top, with McKesson.  I am not sure why products are also included as organizations but I am not an expert in Data Services.

6keythemes.PNG

The above shows top noun groups mentioned and this to me shows some key themes from just a few pages – Mobile and business users.  You can also see real time, use cases, drive user adoption, all key themes this year in the analytics space.

Will you be attending?

To report this post you need to login first.

5 Comments

You must be Logged on to comment or reply to a post.

  1. Susan Keohan

    Hi Tammy,

    I don’t know how you find the time, but I love reading your explorations! 
    No, I won’t be at Reporting 2013, but I hope we hear more from those who do attend.

    Cheers,
    Sue

    (0) 
  2. Ina Felsheim

    Love how Text Analysis helped out here. Would be great if WISPubs would analyze pre, during, and post-event tweets to see if the talked-about topics matches the topics the brochure promotes.

    (0) 

Leave a Reply