Features clustering

Feature clustering (Co-Clustering)

Representation

Keys _id, name, stepId, domainInformation, dataset, project, processingInfo, creationTS, latestUpdateTS are unchanged and follow the classic of representation only dataSpecification keys are changed and are described as above :

• dataSpecification :
• keyword: Its value is “featuresClustering” . Cf to mandatory keys.
• valueType: Cf to mandatory keys.
  • dataType: Its value is “categorical” . Cf to mandatory keys
  • structureType: Its value is “scalar” . Cf to mandatory keys
• meaning: Its value is “feature-clustering” . Cf to mandatory keys.
• view: Cf to mandatory keys.
• dataLocationId: Cf to mandatory keys.

{
  "dataSpecification": {
    "keyword": "columnsClustering",
    "valueType": {
      "dataType": "categorical",
      "structureType": "scalar"
    },
    "meaning": "Columns clustering",
        "view": {
      "id": "637ce534dd85c10875c4fe26",
      "name": "view_11-22-2022_15:05:24"
    },
    "dataLocationId": "62b18d804ae71c6a0025237a",
    "indices": [
      1,
      2
    ]
  }
}

Observation

Key value is an integer describing feature cluster identifier of the given feature (featureId).

{
  "featureId": 0,
  "value": 1,
  "representationId": "hexId",
  "dataLocationId": "62b18d804ae71c6a0025237a"
}

Warning : featureId is not observationId.