Configuring Machine Learning models

Before you begin

Decide which complex scenario you want to address. For instance, to monitor the correlation between temperature, use an HVAC (Heating, Ventilation, and Air Conditioning) system where the objective of HVAC is to monitor the correlation between temperature and humidity under normal operating conditions.

To create a new Machine Learning model

1. On the BMC Helix Edge page, select Machine Learning.
2. On the Machine Learning page, click New Model.
   

3. On the New Model page, enter the following information:

   Field name	Description
   Name	Assign a unique and descriptive name to the Machine Learning model. For example, if your model aims to correlate temperature with humidity, consider naming it Temperature-Humidity-Correlation.
   (Optional) Description	Offer a summary of what the model seeks to achieve. For example, type This model aims to monitor and analyze the correlation between temperature and humidity in HVAC systems.
   Algorithm type	Select Anomaly as the algorithm to be used in the chosen type. This version supports only Anomaly.
   Algorithm	Select HedgeAnomaly from the list to be used for this model. This version supports only HedgeAnomaly.
   Model Information	Confirm the metadata or parameters that pertain to the model. For example, Helix Hedge AutoEncoder is based on an Anomaly detection Algorithm.
   Device profile	Associate the model with a specific device or set of devices. Select the appropriate device profile from the available options to ensure the model is compatible with the target hardware—for example, HVAC (provides humidity and temperature data).
   Model fields	Select the features or variables that the model uses for training. Add or remove fields that serve as input features or target variables for the model. For instance, under Device Attributes, select deviceName, and under Metrics, select Temperature and Humidity.
   Data options	Set the collection interval and add filters to refine the data used for training. Based on recommendations from domain experts or data scientists, select multiple attributes.   In the Collection interval and filters pane, perform the following steps: In the Time period and Sample every field, specify the interval for data collection and apply any necessary filters. For example, in the Time Period field, type 60 days; in the Sample Every field, enter 60 seconds. To filter the data, perform the following steps:  From the Attribute field list, select an attribute. For example, select deviceName.  From the Operator list, select the operation to apply. The available options are Contains, Equals, and Excludes. For example, select Contains. From the Value list, select a value. For example, select HVAC-A. Click +.  Note: When managing filters, it's important to note the following rules regarding operator options, which include EQUALS, CONTAINS, or EXCLUDES: If you use an EQUALS filter for a specific attribute, do not employ CONTAINS or EXCLUDES for that same attribute in another filter. If you opt for CONTAINS as the operator for a filter attribute, use only EXCLUDES for the attribute in another filter, and the converse is true.
   Merge conditions	Use this option when you have multiple device profiles. This option is visible when you use multiple device profiles.
   Preview data	Review the historical data selected for training the model.  Note: This field is visible only after you save the Machine Learning configuration.

4. Click Save.

Where do I go from here

Training-a-machine-learning-model