Need Help?

Luminoso Software Update: 3/1/2016

Last Updated: Jun 20, 2016

Score Drivers Widget and API - Beta (released 12/12/15)

Product

Analytics Dashboards

Clients impacted

Analytics Dashboards customers

 

Change

In Analytics Dashboards, we have added a “Score Drivers” widget.  Given subsets of data that represent quantitative scores (e.g. star ratings, NPS scores, employee engagement levels), the Score Drivers widget calculates which terms are most associated with positive and negative changes in these scores.  

Screenshot(s)

Code snippet (Python)

The documentation for the API endpoint can be found at: analytics.luminoso.com/api/v4/#Drivers-[BETA]

 


New Analytics URL (effective 1/9/16)

Product

Analytics (UI and API)

Clients impacted

All Analytics customers

Backward-compatible?

Yes

Change

We have changed our URLs such that:

 
  • The Analytics UI has moved from dashboard.luminoso.com to analytics.luminoso.com

  • The Analytics API and API documentation has moved from api.luminoso.com to analytics.luminoso.com/api/

 

The previous URLs will redirect to the new URLs and will continue to do so for some time.


Fragment Vectors (effective 2/20/16)

Product

Analytics

Clients impacted

Customers using document vector endpoints (including doc search, topic-subset associations, timelines, prediction/classification)

Backward-compatible?

This will apply to all new projects and any existing projects that are recalculated.  

Change

All terms and documents in our system are represented as 150-dimensional vectors.  Previously, document vectors were calculated by summing the vectors of all the terms and fragments.  (For more information on terms and fragments, see http://support.luminoso.com/customer/portal/articles/1936153)  Going forward, document vectors will be calculated using only the vectors of actual terms within the document.  

 

Previously:

v(“The New York Times sits on the table .”) = v(“New York Times”) + v(“New York”) + v(“New”) + v(“York”) + v(“Times”) + v(“sit”) + v(“table”)

 

Going forward:

v(“The New York Times sits on the table .”) = v(“New York Times”) + v(“sit”) + v(“table”)

More Support

6560c840000970c2e652c21ccafb3167@luminoso.desk-mail.com
http://assets1.desk.com/
false
desk
Loading
seconds ago
a minute ago
minutes ago
an hour ago
hours ago
a day ago
days ago
about
false
Invalid characters found
/customer/en/portal/articles/autocomplete