Free Practice Questions for Fortinet NSE6_FSM_AN-7.4 Exam

The exhibit shows a machine learning Regression configuration. In FortiSIEM regression, selected numerical fields are used as predictors, and another field is selected as the value to predict. The FortiSIEM 7.4 User Guide’s Machine Learning section lists Regression separately and describes the training workflow, including selecting fields used by the algorithm. The exhibit shows Memory Utilization, Sent Bytes, and Received Bytes selected under Fields to use for Prediction, while CPU Utilization is selected under Field to Predict. This means Memory Utilization is an input variable used by the model. If memory utilization is consistently high during training or retraining, that higher value becomes part of the learned model pattern. FortiSIEM does not automatically trigger a high-memory incident merely because a regression input value is high; an incident would require inference behavior and an anomaly action or rule configuration. Option C is also wrong because FortiSIEM does not dynamically lower a CPU trigger threshold just because memory is high. The correct behavior is model update based on the higher observed input value.

The correct answer is A . In FortiSIEM machine learning training, the Train factor controls how much of the prepared dataset is used for training and how much is reserved for testing. The FortiSIEM 7.4 User Guide states in the regression training procedure that the analyst must choose the Train factor and that it “should be greater than 70%.” It also explains the meaning of the value: 70% of the data is used for training and 30% is used for testing. This directly supports option A. Option B is incorrect because the supported running modes include Local, Local Auto, and AWS; “ML” is not the required run mode. Option C is incorrect because regression can use multiple fields for prediction; the guide instructs the user to choose the fields to use for prediction and the field to predict. Option D is incorrect because prediction input fields and target field should not be the same. Regression builds a model to predict a target field based on other fields, so the configuration must use a valid Train factor rather than matching predictor and target fields.

The correct answer is B because the configuration groups results by Source IP and Destination IP , while using COUNT(Matched Events) as a display/aggregate value. FortiSIEM’s grouping logic combines events only when the selected Group By attributes match. The Study Guide explains that Group By attributes determine how matching events are placed into rows, and that when multiple events share the same grouped values, “they are grouped together in one row.” The count column then tracks the number of events represented by that row. In the exhibit, Source IP and Destination IP are the grouping fields, so FortiSIEM displays each unique connection pair once. The count shows how many matching allowed firewall connection events were seen for each pair. Option A is not correct because the exhibit does not show sorting by destination IP hit count. Option C ignores the source and destination grouping. Option D would require grouping by source IP alone or by distinct destination counts per source, which is not the shown configuration.

The correct answer is A because CMDB objects are referenced from the Value field after selecting the appropriate event attribute and operator. The FortiSIEM Study Guide gives a structured search example that references the CMDB. In that example, the attribute is Reporting IP, the operator is IN, and the value is selected from CMDB groups such as Devices: Windows and Networks: Inside Net. The guide explains that to show events reported by Windows servers within a specific network, you set the attribute and operator first, then browse the CMDB and select the relevant CMDB group value. This confirms the workflow: the CMDB reference is chosen as the value of the condition, not as the attribute itself. Option B is incorrect because the CMDB tab is not used to launch the analytics search this way. Option C is not a valid workflow. Option D is wrong because the attribute is selected from event or CMDB attribute lists, while the CMDB object or group is selected in the value field.