General Reference

Classifier Standard List Records

Classifier Standard List records are used to organize and store Standard Lists. This topic provides an alphabetical list and description of the fields that exist for the Classifier Standard List family. The information in the table reflects the baseline state and behavior of these fields.

This family is not enabled for site filtering, which means that records in this family can be accessed by any user with the appropriate license and family privileges. For more information, refer to the Site Filtering section of the documentation.


Field	Data Type	Description	Behavior and Usage
Include in Fuzzy Match?	Boolean	Indicates whether or not the Standard Value is used in fuzzy matches.	When Include in Fuzzy Match? is true, the corresponding Standard Value will be used when a fuzzy logic script is identifying both exact matches and fuzzy matches. When Include in Fuzzy Match? is false, the corresponding Standard Value will only be used when a fuzzy logic script is identifying exact matches. This field is used to improve a fuzzy logic script's ability to predict target values for words that are problematic for fuzzy matching, such as words that form portions of other words (e.g., oil forms a portion of coil) and multiple-word Standard Values (e.g., Radial Bearing).
List Name	Character	The name of the standard list.	Organizes records in the Classifier Standard List into distinct standard lists. The value in this field can be referenced by scripts to identify standard lists for standard values and standard lists for text cleaning.
List Reference	Character	The name of a reference value by which standard lists can be further organized into sub-sets.	Contains values that can be referenced within scripts and cognitions to filter Standard Lists into smaller sub-sets. For example, Compressor, Pump, and Turbine are List Reference values that represent sub-sets of standard values in the maintainable item standard list. The value Generic should be used when not specifying a specific list reference value. When a cognition or script is executed, the list sub-set used is determined as follows: If the value matches a List Reference value, the corresponding sub-set of the standard list is used. If the value does not match a List Reference value, or if no value is specified, the script uses the sub-set Generic. If the value does not match a List Reference value, but is a word defined in the Generic sub-set, the Generic sub-set is used, but the matching word is omitted. For example, in the case of the maintainableItem.py script, this is done so that the equipment’s class is not predicted as a maintainable item of that equipment (e.g., so that valve is not predicted as a maintainable item of a pressure valve).
Standard Code	Character	A code that corresponds to the Standard Value.	None
Standard Value	Character	The unique target value for the standard list.	Contains values for standard lists for standard values and standard lists for text cleaning.
Weight	Numeric	A number representing a weighted value.	Contains an integer of 1 or higher. This value determines which Standard Value a script will use as the prediction when multiple matches are found with the same score. In the case of a tie score, the script will select the word with the higher weight.
Word	Character	A word or term that corresponds to the Standard Value.	Contains words that are mapped to a Standard Value.

Classifier Training History Records

Classifier Training History records store the ratings specified by users when they train a machine learning script (i.e., classifier) in the Test Results workspace of the script. This topic provides an alphabetical list and description of the fields that exist for the Classifier Training History family. The information in the table reflects the baseline state and behavior of these fields.

This family is enabled for site filtering, which means that records in this family can be assigned to a specific site, and will then only be accessible to users who are assigned to the same site and have the appropriate license and family privileges. For more information, refer to the Site Filtering section of the documentation.


Field	Data Type	Description	Behavior and Usage
Classifier	Character	The machine learning script (i.e., classifier) that was trained.	This field is populated automatically.
Prediction	Character	The prediction provided by the classifier.	This field is populated automatically.
Standard List	Character	The standard list used by the classifier.	This field is populated automatically.
Training Date	Date	The time in UTC when the classifier was trained.	This field is populated automatically.
Training Text	Text	The cleaned text that was used by the classifier to determine a prediction.	This field is populated automatically.
Training User	Character	The ID of the user who trained the classifier.	This field is populated automatically.
User Value	Character	The correct value as specified by the user.	When the prediction is a Boolean, the User Value is determined based on the User Vote applied to the Prediction. For example, if the classifier predicted True, and the user votes that the prediction is incorrect, the User Value is specified as False automatically. This field is populated automatically.
User Vote	Boolean	The user’s indication of whether the classifier’s prediction was correct or incorrect.	This value is specified when the user selects or in the Rating column. This field is populated automatically.