Check Anomaly

5min

This model can verify whether a data point is an anomaly, for a specific data point when the data is provided through Qualetics data streaming. Data ingestion through Qualetics API is required for this model to return results.
Note: Data Streaming﻿ integration is required for this model. Please complete the integration setup, verify the data connection﻿ and ensure that the streaming is Live﻿ before using this model.
Model Input Parameters
 Name
 Type
Required
Pre-Defined Values
Info
fieldname
Text
Yes
﻿
Name of the field on which we need to detect the anomalous values.
value
Number
Yes
﻿
Provide a specific value to check whether it is an anomaly
timestamp
DateTime
Yes
﻿
Provide the timestamp corresponding to the value in YYYY-MM-DD HH:MM:SS format
threshold
Number
Yes
0 - 100
What percentage of deviation from the mean should we consider as an anomaly?
smoothness
Text
﻿
low, normal, high
Smoothing or alpha value -  do you want to smoothen the spikes and dips in the data? If not provided, "normal" is applied as the default value 
duration
Number
﻿
﻿
Duration in number of days to fetch data for detecting anomalies. If not provided, a value of 1 is applied as the default value
actor
Text
﻿
﻿
If looking for anomalies in a specific user's data, provide the user id. This value needs to match the Actor id included in the Actor﻿ data object. 
If not provided, all user data is considered.
action
Text
﻿
﻿
If looking for anomalies for a specific type of event, provide the action or event name.  This value needs to match the Action names included in the Action﻿ data object. 
If not provided, all event data is considered.
context
Text
﻿
﻿
If looking for anomalies for a specific context, provide the context name.  This value needs to match the Context names included in the Context﻿data object. 
If not provided, all event data is considered.
Rest API Input Example
The following Rest API code is used to verify whether a particular event is an anomaly based on the values recorded on a temperature sensor data stream, and specifically checking whether deviation from the mean is greater than or equal to 40%.
JS
var settings = {
  "url": "https://mlapi.qualetics.com/api/datamachine/init?id=<DATAMACHINE_ID>",
  "method": "POST",
  "timeout": 0,
  "headers": {
    ...
    //Add Authorization Headers here
    ...
    "Content-Type": "application/json"
  },
   "data": JSON.stringify({
    "fieldname": "temperature",
    "threshold": 40,
    "smoothness": "high",
    "duration": 1,
    "value": "1.22876",
    "timestamp": "2023-12-16 07:14:27",
    "actor": "sensor1"
  }),
};

$.ajax(settings).done(function (response) {
  console.log(response);
});
var settings = {
  "url": "https://mlapi.qualetics.com/api/datamachine/init?id=<DATAMACHINE_ID>",
  "method": "POST",
  "timeout": 0,
  "headers": {
    ...
    //Add Authorization Headers here
    ...
    "Content-Type": "application/json"
  },
   "data": JSON.stringify({
    "fieldname": "temperature",
    "threshold": 40,
    "smoothness": "high",
    "duration": 1,
    "value": "1.22876",
    "timestamp": "2023-12-16 07:14:27",
    "actor": "sensor1"
  }),
};

$.ajax(settings).done(function (response) {
  console.log(response);
});
﻿
Model Output Result
Parameter Name
Parameter Type
Info
expected value
Number
﻿
deviation from expected
Number
﻿
value
Number
﻿
timestamp
DateTime
﻿
id
Number
Unique ID of the Event
Rest API Output Example
JSON
{
    "deviation from expected": 0.2805779525,
    "expected value": 0.9455036106,
    "fieldname": "temperature",
    "final result": "Anomaly Found",
    "id": 979752,
    "result": "Anomaly Found",
    "sessionid": "d3c76a85-5693-40b2-8313-306f3d34cbce",
    "timestamp": "2023-12-16 07:14:27",
    "value": "1.22876"
}
{
    "deviation from expected": 0.2805779525,
    "expected value": 0.9455036106,
    "fieldname": "temperature",
    "final result": "Anomaly Found",
    "id": 979752,
    "result": "Anomaly Found",
    "sessionid": "d3c76a85-5693-40b2-8313-306f3d34cbce",
    "timestamp": "2023-12-16 07:14:27",
    "value": "1.22876"
}
﻿
Standard Output Parameters
Every model execution output consists of the following standard output parameters
input 
The input string required for the model to extract the categories
original input
This is the input provided to the first step in model which is retained across multiple steps in a Data Machine workflow. 
final result
The result of the model executed in the final step of the Data Machine workflow
sessionid
A unique session id that is generated for every execution of a Data Machine which can be used to retain results across multiple sessions
status
The result of the Data Machine execution. If all of the steps in a sequence are successfully executed, a value of "Completed" is provided.  If the execution is interrupted at any point, a value of "Terminated" is provided with the reason for Termination.
﻿

Name	Type	Required	Pre-Defined Values	Info
fieldname	Text	Yes		Name of the field on which we need to detect the anomalous values.
value	Number	Yes		Provide a specific value to check whether it is an anomaly
timestamp	DateTime	Yes		Provide the timestamp corresponding to the value in YYYY-MM-DD HH:MM:SS format
threshold	Number	Yes	0 - 100	What percentage of deviation from the mean should we consider as an anomaly?
smoothness	Text		low, normal, high	Smoothing or alpha value - do you want to smoothen the spikes and dips in the data? If not provided, "normal" is applied as the default value
duration	Number			Duration in number of days to fetch data for detecting anomalies. If not provided, a value of 1 is applied as the default value
actor	Text			If looking for anomalies in a specific user's data, provide the user id. This value needs to match the Actor id included in the Actor data object. If not provided, all user data is considered.
action	Text			If looking for anomalies for a specific type of event, provide the action or event name. This value needs to match the Action names included in the Action data object. If not provided, all event data is considered.
context	Text			If looking for anomalies for a specific context, provide the context name. This value needs to match the Context names included in the Contextdata object. If not provided, all event data is considered.

Parameter Name	Parameter Type	Info
expected value	Number
deviation from expected	Number
value	Number
timestamp	DateTime
id	Number	Unique ID of the Event

List Anomalies

Semantic Search