Mixed data

Mixed data is a heterogeneous type which consist int combination of multiple regular data types such as numerical and binary data.

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 “mixedData” . Cf to mandatory keys.
  • valueType: Cf to mandatory keys.
    • dataType: Its value is “mixed” . Cf to mandatory keys.
    • structureType: Its value is “vector” . Cf to mandatory keys.
    • columnsType : Columns type description which either “irregular” or (when it will come).
  • meaning: Its value is “Mixed feature type data” . Cf to mandatory keys.
  • view: Cf to mandatory keys.
  • dataLocationId: Cf to mandatory keys.
  • columnsTemplate : It is an JSON Array of JSON Dictionary with 2 keys :
    • id : Identifier of a given column template.
    • value : JSON Array of dictionaries having the keys columnName and columnType respectively describing feature name and type.
      • columnName : String describing the name of the feature.
      • columnType : String describing the type of the feature, it is an enumeration of following types :
        • boolean, float, double, string, integer, long. Missing ones will be completed when needed.

{
  "dataSpecification": {
    "keyword": "mixedData",
    "valueType": {
      "dataType": "mixed",
      "structureType": "vector",
      "columnsType": "irregular"
    },
    "meaning": "Mixed feature type data",
    "view": {
      "id": "637ce534dd85c10875c4fe26",
      "name": "view_11-22-2022_15:05:24"
    },
    "dataLocationId": "62b18d804ae71c6a0025237a",
    "columnsTemplate": [
      {
        "id": 1,
        "value": [
          {
            "columnName": "col1",
            "columnType": "boolean"
          },
          {
            "columnName": "col2",
            "columnType": "float"
          }
        ]
      },
      {
        "id": 2,
        "value": [
          {
            "columnName": "col1",
            "columnType": "float"
          },
          {
            "columnName": "col2",
            "columnType": "string"
          }
        ]
      }
    ]
  }
}   

Observation

Observation keys description :

• observationId : cf definition
• value : JSON Array of String which will be converted accordingly with type info in columns key, it can be predefined in the representation or it can be set in a custom way in the columns key.
• columns : Define if the pattern of this observation follow an existing template defined in the representation or a custom one.
  • type : The String “custom” or “template” depending use case.
  • value : Two scenario :
    • For “template” case the value takes the Integer value which identify a specific template.
    • For “custom” columns name and types are described in the same way as in representation.
• representationId : cf definition
• dataLocationId : cf definition
• source : Data source file.

Custom observation JSON example :

{
  "observationId": 0,
  "value": [
    "True",
    "12.456"
  ],
  "columns": {
    "type": "custom",
    "value": [
      {
        "columnName": "col1",
        "columnType": "boolean"
      },
      {
        "columnName": "col2",
        "columnType": "float"
      }
    ]
  },
  "representationId": "62bda1197750ec088f984461",
  "dataLocationId": "62b18d804ae71c6a0025237a",
  "source": {
    "filepath": "NAS://hephia/file1.mat"
  }
}

Template observation JSON example :

{
  "observationId": 0,
  "value": [
    "True",
    "12.456"
  ],
  "columns": {
    "type": "template",
    "value": 1
  },
  "representationId": "62bda1197750ec088f984461",
  "dataLocationId": "62b18d804ae71c6a0025237a",
  "source": {
    "filepath": "NAS://hephia/file1.mat"
  }
}