Free Fortinet NSE6_FSM_AN-7.4 Practice Exam with Questions & Answers

Grouping by User, Source IP, and Count means that each unique combination of those three attributes will be treated as a separate result. In the table, all six rows have distinct combinations of User, Source IP, and Count - so FortiSIEM will display 6 results.

Six because grouping by User , Source IP , and Count creates a separate result for every unique combination of those three selected attributes. The FortiSIEM Study Guide explains this grouping behavior in the single-subpattern rule example: “If multiple VPN login failure events have the same source IP address, reporting device, reporting IP address, and user, they are grouped together in one row, and the count column tracks the number of events for each of those rows.” Applying that rule here, FortiSIEM compares all selected Group By fields together. In the exhibit, every row has a unique Source IP address, even where the same user appears more than once. For example, Mike appears twice, but the Source IP and Count values are different. Alice appears twice with Count 2, but the Source IP values are different. Bob appears twice, but both Source IP and Count are different. Since no row has the same User, Source IP, and Count combination as another row, FortiSIEM displays all six rows.

The search fails because nested analytics queries require the outer query attribute type to match the inner query display-column data type. The FortiSIEM Study Guide explicitly explains this rule in the Nested Query section: “Another important aspect to understand is that the data value types must match.” It further explains that each inner query display column has a data value type, such as IP, string, or integer, and the outer query attribute must match that type. The FortiSIEM 7.4 User Guide states the same operational requirement: for a nested query to work correctly, “the data type of the filter attribute in the outer query must match up with the data type of one certain display column in the inner query.” Therefore, if the inner query returns a string attribute but the outer query compares it against an IP-type attribute, FortiSIEM cannot produce a valid match. The issue is not the time range, not the use of User and Event Type together, and not the Boolean operator. It is an attribute type mismatch between the query and subquery.

The best answer is A. Two . The exhibit shows an Analytics Search scatter plot with two visually distinct groups of data points: one isolated group around the lower-left area of the chart and another dense group near the upper-right area. For KMeans clustering, the analyst must provide the number of clusters based on the observed structure of the data. The FortiSIEM 7.4 User Guide describes KMeans as “an unsupervised clustering algorithm that groups data points into user specified K groups so that each data point belongs to one group.” It also states that KMeans “tries to iteratively minimize intra-cluster distance and maximize inter-cluster distance” and notes that the user must “specify the number of clusters based on user’s knowledge of data.”

The same FortiSIEM guide explains that during Clustering Local Mode training, the analyst chooses the algorithm and, “for KMeans choose the cluster size as a guess.” In this exhibit, the natural guess is 2, because the scatter plot separates into two obvious groups. Fifty or 100 clusters would overfit the small number of visible groups, while one cluster would merge two clearly separate behaviors into a single cluster.

The correct answer is D. The exhibit shows an outer event query using the Event Attribute filter Source IP NOT IN Device IP: Approved Devices. The outer query time range is set to Relative — Last 10 Minutes, so FortiSIEM searches only the event data from the last 10 minutes. The exhibit also shows a separate Nested Time Range set to Relative — Last 30 Days. In FortiSIEM nested searches, the nested time range applies to the inner report/subquery, not to the outer event search. The FortiSIEM 7.4 User Guide states that nested query functionality lets one query refer to results from another query, and for outer event / inner event nested searches, it instructs the user to “choose the time range for outer query” and separately “choose Nested Time Range for the inner query.” It also states that when an existing query is used as an inner query, “time range would be set separately” in the outer query configuration. Therefore, FortiSIEM searches the last 10 minutes of outer events and compares their Source IP values against the Device IP values returned by the Approved Devices report using the last 30 days nested time range.

FortiSIEM grouping works by combining events that have the same values for all selected Group By attributes. The Study Guide’s rule and subpattern example states that when multiple events have the same Group By values, “they are grouped together in one row, and the count column tracks the number of events for each of those rows.” In the exhibit, the selected grouping fields are Reporting Device, Reporting IP, and Application Category. The table contains six raw rows, but two rows share the same grouped combination: FW01 / 10.1.1.1 / DB . Those two rows are collapsed into one grouped result. The other combinations are unique: FW02 / 10.1.1.2 / WebApp, FW01 / 10.1.1.1 / SSH, FW03 / 10.1.1.3 / DB, and FW04 / 10.1.1.4 / SSH. That creates five grouped rows in total. FortiSIEM does not display six because grouping removes duplicate combinations, and it does not display four because only one duplicate combination exists.

The best matching analytics search is Username CONTAIN smit. The FortiSIEM Analytics lesson explains keyword and attribute matching by using the CONTAIN operator for searching values inside raw logs or attributes. The Study Guide gives a direct example of FortiSIEM translating keyword logic into a condition such as “Raw Event Log CONTAIN TCP.” That shows the operator is designed for substring-style matching rather than full-value equality. For a user such as JSmith, the string fragment smit is contained inside the username value, so the condition can match events where the username appears as JSmith, jsmith, or possibly in a longer identity string such as a domain-qualified username. Option A is wrong because smith is not the complete username and uses an equality-style comparison. Option B is wrong because it excludes usernames ending in jsmith. Option C uses the wrong operator form for this context. Therefore, the practical FortiSIEM analytics condition that can match the target failed-login event is Username CONTAIN smit.

The verified answer is D. Five . FortiSIEM grouping is based on unique combinations of the selected Group By fields. The Study Guide explains this behavior clearly: if multiple events have the same selected Group By values, “they are grouped together in one row,” and the count column tracks the number of events for each row. In this question, the selected fields are User and Count . The six raw rows contain these combinations: Mike/4, Bob/3, Alice/2, Alice/2, Bob/6, and Mike/5. Because Alice/2 appears twice, those two rows are grouped into a single result. The remaining combinations are unique. So FortiSIEM displays five grouped results, not six. Six would be correct only if every row had a unique User-and-Count combination, or if grouping included another differentiating attribute such as Source IP. Since the question specifically groups only by User and Count, duplicate User/Count pairs collapse into one row. Therefore, the correct result count is five .

The correct answer is B. Train. In FortiSIEM machine learning, the Train step is where the model is built from the prepared dataset and where model-fitting behavior can be adjusted. The FortiSIEM 7.4 User Guide explains that after preparing data, the analyst goes to Analytics > Machine Learning > Train, selects the machine learning task, chooses the algorithm, selects the prediction/target fields when required, and chooses the Train factor, which determines how much data is used for training versus testing. The guide states that the Train factor should be greater than 70%, meaning 70% of the data is used for training and 30% for testing. It also explains that model quality metrics show how accurately the algorithm predicts the field. For regression, lower MAE means a better fit, and R2 shows how well predictions approximate real data points. Most importantly, the guide states: “If you want to change the algorithm parameters and re-train, then click Tune & Train, change the parameters and click Save & Train.” This confirms that modifying how the model fits the training dataset is done in the Train step, not Prepare Data, Statistics, or Design.

The correct answer is Local Auto. The uploaded answer was right, but its explanation was sloppy because it incorrectly described Local mode as the most time-consuming mode. In FortiSIEM machine learning, Local Auto mode selects the best algorithm by evaluating multiple candidate algorithms. The User Guide states that in Local Auto mode, “FortiSIEM picks the best algorithm” and that the Max Run Time parameter limits how long the job can run; longer runtime can produce better results. That is why Local Auto can take the most time. Forecasting and Regression are task types, not run modes.

The FortiSIEM Study Guide defines a nested query as a query that has another query embedded within it, where the embedded query is known as the subquery. The FortiSIEM 7.4 User Guide lists the supported nested search scenarios: Outer CMDB Query, Inner Event Query ; Outer Event Query, Inner Event Query ; and Outer Event Query, Inner CMDB Query . These supported combinations prove that the inner or subquery can be either an Event Query or a CMDB Query. LDAP Query and SNMP Query are not listed as supported nested analytics subquery types. LDAP may be used for user discovery or authentication, and SNMP may be used for monitoring or discovery, but neither is referenced as a nested analytics query type. The User Guide also explains that the outer query uses Select from Report to reference the saved inner query and that the matching attribute data types must align between the outer query and the selected display column in the inner query.