Sample Document
Sensible's query language, SenseML, exposes dozens of document parsing primitives powered by ML, NLP, and good old-fashioned rules and heuristics
No new code, just new JSON configuration
Our full-automation platform doesn't rely on human-in-the-loop, which means you don't have to wait around for results
We handle OCR, deskewing, ligatures, text cleaning, line merging, and all the other headaches of the PDF format so that you don't have to
{
"fields": [
{
"id": "case_type",
"method": {
"id": "box"
},
"anchor": "case type"
},
{
"id": "class",
"method": {
"id": "label",
"position": "right"
},
"anchor": "class"
}
]
}
Our simple, expressive DSL empowers operations staff and other non-technical users to onboard documents, without having to write any code
{
"fields": [
{
"id": "effective_date",
"type": "date",
"method": {
"id": "label",
"position": "right",
"tiebreaker": "<"
},
"anchor": {
"match": {
"text": "Policy Period:",
"type": "startsWith"
}
}
}
]
}
Use our powerful NLP primitives to summarize and classify free text, fuzzy match tables and key/value pairs, and filter documents down to just their relevant sections
{
"fields": [
{
"id": "construction_invoice",
"type": "table",
"method": {
"id": "invoice",
"columns": [
{
"id": "quantity",
"terms": [
"qty",
"quantity",
"ordered",
"shipped"
]
}, {
"id": "description",
"terms": [
"description",
"product"
]
}
]
}
}
]
}
Get up and running immediately with one our pre-built parsers for common forms
