Feedback on NI Community

cancel
Showing results for 
Search instead for 
Did you mean: 

The LIST

Here is a 6.1 version, dedicated to shoneill...

CC
Chilly Charly    (aka CC)
Message 101 of 131
(1,583 Views)
Hi,

I had a problem with Excel Get Data Specified Field.vi, it didn't like the Value property. Had to change it for the only valid option which was value2.

I'm using Office 2003. LabVIEW 7.1

It seem to run and produces a lot of dots, not sure what its trying to convey.

regards
Ray Farmer

Message Edited by Ray Farmer on 06-01-2005 10:02 PM

Regards
Ray Farmer
Message 102 of 131
(1,579 Views)
No Problems with Office 2000 and LV 7.1.

Need some explanation about the meaning of the axis too.
Waldemar

Using 7.1.1, 8.5.1, 8.6.1, 2009 on XP and RT
Don't forget to give Kudos to good answers and/or questions
Message 103 of 131
(1,563 Views)
Some explanation about the meaning of the 3 axis.
As explained in a previous post, the axis are automatically generated, as a linear combination of all the adressed variables. The attached picture shows a 2D representation of the position, in the new coordinates, of the original variables.

Three directions can be defined :
The first one, pointing to the right, is defined by the number of threads (as % of the overall contributions), and the number of attachments (as % of posts with a document attached);
The second direction is exactly opposite, and defined by the number of replies (as % of the overall contributions), the average rating, the number of URL (as % of posts with an URL attached) and the ratio ratings/replies ;
The third direction is orthogonal to the two others, pointing downward, and defined by the number of replies/day during the last month and since the inscription date, the total number of replies, and the % breaks.

The interpretation is rather straight forward :
The first direction correspond to the forum users that are mainly consumers, since they have a high % of threads and attachments;
The second direction corresponds to the Enthusiasts whith a high % of replies.
As a matter of fact, this score is well correlated with a high average rating, a high % of URL attached, and a high ratings/replies ratio.
The third direction corresponds to the Enthusiasts activity.

Now, if you have a look at the way the points are distributed in the 3D graph, you will see that they belong essentially to two secant planes, roughtly parallel to the third axis. If you look from above, in the direction of Axis 3, you will find that the points, projected on the Axis 1 & 2 plane, are disposed along the directions that we have just discussed.
It is very easy to spot the most prominent Enthusiasts, and the most prominent consumers. Most of the List members are essentialy located within + or -1 from the graph origin (standard deviation).

Accordingly, Axis 1 gives the type of activity (questions against replies) and Axis 2 measures the intensity of the activity.
The interpretation of the third axis is more difficult, but it seems to differntiate the enthusiasts giving replies with URL attached, from those who answer with vis attached.

Not sure these explanations are cristal clear. The discussion is opened...

CC
Chilly Charly    (aka CC)
Message 104 of 131
(1,628 Views)
Additionnal infos, for those who are not used to 3D graph manipulations :
- the graph can be rotated using the mouse : click and draw
- you can zoom in and out, using the wheel, after a click in the graph pane,
or with Alt + click & draw
- you can move the graph with shift + click & draw.

CC
Chilly Charly    (aka CC)
Message 105 of 131
(1,627 Views)
Hum, my first post was so long that I forgot the attachment :

Message Edited by chilly charly on 06-02-2005 11:59 PM

Chilly Charly    (aka CC)
Message 106 of 131
(1,625 Views)
Hi CC,

Let me get this right. You are attempting to start a conversation where you introduction to the subject uses the word "orthogonal"?

This will definately have to wait until the week-end.

Ben
Retired Senior Automation Systems Architect with Data Science Automation LabVIEW Champion Knight of NI and Prepper LinkedIn Profile YouTube Channel
0 Kudos
Message 107 of 131
(1,604 Views)
Hi CC,

I had a chance to look at this closer and just had to change things (:<)) .

Using a log scale on all of the axis breaks the cloud up nicely into a couple of clusters.

I also color coded the points based on the magnitude of the vector. So what's your favorite color CC?

The new VI is "Enthusiasts Log 3D PCA.vi". Make sure your previous version is closed before opening this one. I hacked a bunch of stuff.

It looks like we break down as;

Buissnes only
Mixed
Active
Users

I have to go back over your intro to get a better feel for what I am seeing.

Cool stuff CC!

Ben
Retired Senior Automation Systems Architect with Data Science Automation LabVIEW Champion Knight of NI and Prepper LinkedIn Profile YouTube Channel
0 Kudos
Message 108 of 131
(1,576 Views)
Ben,

I' missing the "Color from magnitude.vi". Is this part of a toolkit I haven't installed?

And can you take into account that there are people which are living with a lower resolution than others. I use 1152*864 and cannot see the 3D-Graph complete.

The log axis was a good idea.

Message Edited by waldemar.hersacher on 06-04-2005 10:06 PM

Waldemar

Using 7.1.1, 8.5.1, 8.6.1, 2009 on XP and RT
Don't forget to give Kudos to good answers and/or questions
0 Kudos
Message 109 of 131
(1,590 Views)
Sorry, did not save to llb.

Ben
Retired Senior Automation Systems Architect with Data Science Automation LabVIEW Champion Knight of NI and Prepper LinkedIn Profile YouTube Channel
0 Kudos
Message 110 of 131
(1,592 Views)