Tag: Biostatistics

Customising the Configured thresholds for predictive analysis

Symantec Community Leave a comment
I need a solution

Hi;

The current thresholds for Cylence are 4 and 7. With 7 and above is the threshold for a block and 4 and below is the threshold for a good benign file. My understanding is that anything in between 4 and 7 will be sent to the AV engine to be scanned. Would it make more sense to avoid any false positives and negatives to change these limits to 3 and 8. Are the default limits the recommended ones and if so, is there a reference I can use for that.

Kindly

Wasfi

0

1566430266

Related:

  • No Related Posts

Once I calculate the trimmed means using the syntax provided on this site, can I use the upper bound and lower bound confidence intervals to delete extreme values in my dataset and then re-analyze the data?

IBM Customer Leave a comment
Once I calculate the trimmed means in SPSS using the syntax provided on this site, can I use the upper bound and lower bound confidence intervals to delete extreme values in my dataset and then re-analyze the data?

Related:

Particular coordinates on an ROC in SPSS

IBM Customer Leave a comment
I’ve created a predictive score by rounding the beta coefficients from multivariable logistic regression to whole numbers and summing for the total score (SPSS). When I look at the ROC for the score, the coordinates are given as .5, 1.5, 2.5 etc which isn’t helpful for determining the sensitivity/specificity of a particular score (always a whole number). Is there a way I can specify to report only whole numbers for coordinates or work out the sensitivity/specificity from a certain score, i.e. score 2?

Related:

Correlations Bootstrapping Output Missing

IBM Customer Leave a comment
I want to run a Pearson’s r with bootstrapped BCa 95% confidence intervals. I select the options, however, the output is a normal correlation matrix missing bootstrap information. Figures attached. Is this is bug with version 25?![alt text][1]

[1]: /answers/storage/temp/17461-problem.png

Related:

Graphs not apprearing in output viewer

IBM Customer Leave a comment
I am using SPSS v24.0, have run binary logistic regression and multilevel binary logistic regression (genlinmixed) models, saved predicted probabilities from models I want to compare, and run ROC Curve analysis using the predicted probabilities. I’m working with an extremely large dataset, but the analysis runs perfectly well (if a bit slowly) — however, the ROC curve graphs display as ‘Processing…’ and then never show up. I’ve left this sitting overnight in case the graphs are just loading slowly, but no luck.
Even more strangely, now other graphs that used to display properly seem to be showing the same problem (e.g. graphs for specific effects at the end of a multilevel analysis output from genlinmixed syntax).
Note: I have not changed any settings since before this problem appeared, nothing new has happened with/to my desktop computer, and the only new factor is that I hadn’t run the ROC Curve command before.

Related:

Difference between XForce remote networks and “IP categorized by XForce”

IBM Customer Leave a comment
Hi,

Are there any difference between the following rule tests(except for the confidence)?
1. destination IP is a part of XForce.Premium.Premimum_Malware
2. destination IP is categorized by X-Force as Malware with confidence value greater than 69

Are the IPs in XForce.Premium.Premium_Malware exactly the same as IPs in option 2 above?
On v7.2.8, what is the update interval for XForce.Premium.Premimum_Malware?
Do the “Enhanced XForce Rules” and “Legacy Xforce Rules” use the DB?

Related: