Skip to Content

Creating a Machine Learning model by merging data

Merging data from multiple device profiles refers to aggregating data and information from different devices into a unified and coherent view. 

This approach leverages information from various sources to create a more comprehensive and accurate predictive model.

Information
Scenario

You manage BMC Helix Edge deployment for a telecommunications company and have various device profiles representing different aspects of your Telco tower infrastructure. These device profiles include the following metrics:

  • Call Volume (BSS profile): This data monitors call traffic and usage on the Telco tower.
  • Fault (OSS profile): This data detects any faults or issues in equipment and systems.
  • Antenna Dislocation (Telco Tower profile): This data tracks the positioning and status of antennas on the tower.

You can merge data from these different device profiles to detect and address potential issues that might arise due to a correlation between call volume, tower faults, and antenna dislocations. You analyze the merged dataset to identify patterns or correlations.

For example, you might observe the following issues:

  • A sudden increase in call volume coinciding with a fault alert, therefore indicating a potential problem.
  • Antenna dislocations during high call volume, which leads to service disruptions.

Before you begin

Make sure you have a device profile from the available options to make sure the model is compatible with the target hardware.

  1. On the BMC Helix Edge page, select Machine Learning.
  2. On the Machine Learning page, click New Model.
    BMC Helix Edge displays the following screen:
    image-2024-2-5_14-52-50.png

  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 in the chosen type. This version supports only Anomaly.  

    Algorithm

    Select HedgeAnomaly for this model from the list. This version supports only HedgeAnomaly.

    Model Information

    Displays additional metadata or parameters that pertain to the model. For example, Hedge AutoEncoder is based on an Anomaly detection Algorithm.

  4. From the Device profile list, select the appropriate device profile from the available options to make sure the model is compatible with the target hardware.
    Make sure you select a profile with two or more     devices. Merging conditions apply when you select more than one profile. You must select devices that exhibit a logical connection based on their relevance, determined by area, location, or tag.
  5. In the Model fields pane, click Select to select attributes and metrics. 
    BMC Helix Edge displays the following Select model fields screen:
    image-2023-12-13_16-58-50.png
  6. To merge data from other profiles, perform the following steps to select the attributes and metrics that the model uses:

      1. Click image-2023-12-13_17-21-11.pngto enable the Merge data from other profiles options.
        BMC Helix Edge displays the following option to add attributes and metrics from multiple profiles:
        image-2023-12-13_17-24-5.png

        Information
        Important

        BMC Helix Edge displays the Merge data from other profiles button at the bottom of the screen only when two or more profiles share identical attributes or context fields within their profile definitions. For example, data from a profile named TelcoTower with an attribute called siteId and data from another profile named OSS, also with the siteId attribute, can be merged.

      2. To select device attributes and metrics for a device profile, click > to expand the options and select the attributes and metrics.
      3. To select device attributes and metrics for another device profile, click > to expand the options and select the attributes and metrics.
      4. Click OK.

  2. In the Data Option pane, enter the following information:

    Field name

    Description

    Collection interval and filters

    Set the collection interval and add filters to refine the data used for training. We select multiple attributes based on best practices from domain experts or data scientists.

    image-2023-12-13_16-15-22.png

    • To specify the interval for data collection, set the Time period and Sample every field. For example, to train the collection time, enter 60 days in the Time Period field and 60 seconds in the Sample Every field.
    • To apply filters:
      1. In the Attribute list, select an attribute to focus on. 
        For example, select deviceName. 
      2. In the Operator list, select the operation to apply. The available options are Contains, Equals, and Excludes.
        For example, select Contains.
      3. In the Value field, enter a value. 
        For example, select HVAC-A.
      4. Click + to add the filter. 
        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.

    BMC Helix Edge displays the following pane with merge criteria: 
    image-2023-12-13_17-48-15.png

  3. In the Merge conditions pane, select the criteria from the available options. 
  4. Click Save.

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*

BMC Helix Edge 24.3