• stefanconrady
  • Administrator
  • Email
  • Posts 8
  • Reputation 0
  • Member Since
  • Last Active
  • Name Stefan Conrady
  • Gender Male
  • Location Franklin, TN
All Posts Topics Started Likes
BayesiaLab Integration with Other Applications
Q.  Is it possible to integrate BayesiaLab with other applications..?

A. Yes, the best way to integrate BayesiaLab in other applications is using the Bayesia Engine API: http://www.bayesia.com/bayesia-engine-api
0 0
Equipment Diagnosis with Bayesian Networks
Q. Do you have examples of systems and/or equipment diagnosis, or other examples out of the healthcare domain?  If not, is this something you might consider for future webinars?

A. Equipment diagnosis was one of the earliest applications of Bayesian networks. We haven't done a seminar or webinar on that topic, but we had a presentation by the U.S. Army on that topic at our BayesiaLab Conference in Los Angeles in 2013. We'll look for the video and then share the link.

You may also want to check out the BEST application, which focuses exclusively on equipment diagnosis: http://www.bayesia.com/best-overview
0 0
Structural Coefficient
Q. I thought the Structural Coefficient was the weighting factor on Complexity? So 0.4 means a less complex model with respect to the data fit?

A. You can consider the Structural Coefficient 𝛼 a "penalty weight." The MDL score is computed as follows: MDL(B,D)=𝛼⋅DL(B)+DL(D|B). So, lowering 𝛼 below 1 increases the complexity of the network.

Please also see the BayesiaLab Library in this regard: http://library.bayesia.com/display/BlabC/Structural+Coefficient
0 0
Acceptance Threshold & ROC Curve

Q. Is the Acceptance Threshold the ROC cutpoint?

A. Yes, you could use the ROC curve to identify the optimal cutpoint, which you would then use as the Acceptance Threshold.

0 0
Confidence Intervals
Q. Q: Is there a way to include the amount of data used to make the estimate of probability? I feel like there should be more uncertainty associated with an estimate based on fewer data points than one with more even with the same probability

A. Yes, there is. Please see this page in the BayesiaLab Library for information on the Confidence Interval Report: http://library.bayesia.com/pages/viewpage.action?pageId=18350258
0 0
Interactions and Nonlinearities
Q. Q. Can BayesiaLab discover interactions and discover nonlinearities?

A. Absolutely. While we haven't explored these types of relationships in detail in the webinar, that's one of the key applications of BayesiaLab.
0 0
GSS Survey Design?
Q. Were the survey questions designed with the distinction between liberals and conservatives in mind?

A. No, the survey questions were very generic and covered a broad spectrum of topics. You can learn more about the General Social Survey here: http://gss.norc.org/
We simply took the existing survey and tried to find segments of opinions, just like we would do in a segmentation exercise for marketing purposes
0 0
This forum
Hi Joe, we migrated the old forum to a new platform, and in that context we ended up with some gobbledygook in existing posts. We'll try to fix it. Cheers!
0 0
count post selected

Add a Website Forum to your website.