AWS CloudWatch Alarm triggering wrongly - amazon-web-services

Question

I am looking forward to understand how the alarms are triggered.
The case is a follows:
Currently, I have set up a custom metric for a Log Group. These are the logs of an R script running with AWS ECS, and the custom metric is set to 1 whenever it finds the keyword done in the logs, otherwise, it will set to 0.
The alarm was setup to check 1 data point, and whenever it finds 0 it will be triggered.
Now the issue is that I have been receiving the alarm randomly without reason, as the logs indicate the script has been running successfully (All of them contain the keyword done). Then after a while, and before the time period setup happens again, the status changes back to OK.
{
"MetricAlarms": [
{
"AlarmName": "XXXXXXXXX",
"AlarmArn": "XXXXXXXXX",
"AlarmConfigurationUpdatedTimestamp": "2020-07-27T07:08:34.498000+00:00",
"ActionsEnabled": true,
"OKActions": [],
"AlarmActions": [
"XXXXXXXXX:slack-monitoring"
],
"InsufficientDataActions": [],
"StateValue": "OK",
"StateReason": "Threshold Crossed: 1 out of the last 1 datapoints [1.0 (26/07/20 07:10:00)] was not less than or equal to the threshold (0.0) (minimum 1 datapoint for ALARM -> OK transition).",
"StateReasonData": "{\"version\":\"1.0\",\"queryDate\":\"2020-07-27T07:10:51.284+0000\",\"startDate\":\"2020-07-26T07:10:00.000+0000\",\"statistic\":\"Maximum\",\"period\":86400,\"recentDatapoints\":[1.0],\"threshold\":0.0}",
"StateUpdatedTimestamp": "2020-07-27T07:10:51.285000+00:00",
"MetricName": "XXXXXXXXX",
"Namespace": "XXXXXXXXX",
"Statistic": "Maximum",
"Dimensions": [],
"Period": 86400,
"EvaluationPeriods": 1,
"DatapointsToAlarm": 1,
"Threshold": 0.0,
"ComparisonOperator": "LessThanOrEqualToThreshold",
"TreatMissingData": "missing"
}
],
"CompositeAlarms": []
}