Claims loss run example

The following example shows extracting repeated fields from a section containing a list of claims:

• It captures Sept and Oct claims by specifying sections that start with "claim number" and collection range that starts with "September" and ends with "November".
• From each section, it excludes intervening text that isn't part of a section but that does repeat (monthly_number_unprocessed_claims) using a Stop parameter. Instead, it captures this info outside of sections with "match:all".

Config

{
  "fields": [
    {
      "id": "total_unprocessed_claims",
      "type": "number",
      "anchor": "total unprocessed claims",
      "method": {
        "id": "row"
      }
    },
    {
      "id": "monthly_number_unprocessed_claims",
      "match": "all",
      "anchor": {
        "match": {
          "type": "includes",
          "text": "unprocessed claims:",
          "isCaseSensitive": true
        },
        "end": {
          "text": "total claims",
          "type": "startsWith"
        }
      },
      "method": {
        "id": "row",
        "position": "right",
        "includeAnchor": true
      }
    }
  ],
  "sections": [
    {
      "id": "unprocessed_sept_oct_claims",
      "range": {
        "anchor": {
          "start": {
            "text": "September",
            "type": "startsWith",
            "isCaseSensitive": true
          },
          "match": {
            "type": "includes",
            "text": "claim number"
          },
          "end": {
            "type": "startsWith",
            "text": "November",
            "isCaseSensitive": true
          }
        },
        "stop": {
          "type": "includes",
          "text": "unprocessed claims:",
          "isCaseSensitive": true
        }
      },
      "fields": [
        {
          "id": "claim_number",
          "type": "number",
          "anchor": {
            "match": {
              "type": "startsWith",
              "text": "Claim number:",
              "isCaseSensitive": true
            }
          },
          "method": {
            "id": "label",
            "position": "right"
          }
        },
        {
          "id": "phone_number",
          "type": "phoneNumber",
          "anchor": {
            "match": {
              "type": "includes",
              "text": "Phone number",
              "isCaseSensitive": true
            }
          },
          "method": {
            "id": "row",
            "position": "right"
          }
        }
      ]
    }
  ]
}

Example document

The following image shows the data extracted by this config for the following example PDF:

Output

{
  "total_unprocessed_claims": {
    "source": "5",
    "value": 5,
    "type": "number"
  },
  "monthly_number_unprocessed_claims": [
    {
      "type": "string",
      "value": "Sept unprocessed claims: 2"
    },
    {
      "type": "string",
      "value": "Oct unprocessed claims: 1"
    },
    {
      "type": "string",
      "value": "Nov unprocessed claims: 2"
    }
  ],
  "unprocessed_sept_oct_claims": [
    {
      "claim_number": {
        "source": "1223456789",
        "value": 1223456789,
        "type": "number"
      },
      "phone_number": {
        "type": "phoneNumber",
        "source": "512 409 8765",
        "value": "+15124098765"
      }
    },
    {
      "claim_number": {
        "source": "9876543211",
        "value": 9876543211,
        "type": "number"
      },
      "phone_number": null
    },
    {
      "claim_number": {
        "source": "6785439210",
        "value": 6785439210,
        "type": "number"
      },
      "phone_number": {
        "type": "phoneNumber",
        "source": "505 238 8765",
        "value": "+15052388765"
      }
    }
  ]
}