01
OCR, NLP, Computer Vision & generative AI
The unparalleled performance of GPT, for all your documents.
Document capture, recognition, data extraction, verification, and fraud detection have been radically transformed by the latest Artificial Intelligence models.
To shed light on the reality of existing solutions, we ran a test protocol with real data on the latest solutions from Kofax, Abbyy, Nanonets, and others.
The combination of the latest OCR models, Transformers, and generative AI demonstrates its unmatched performance.
To shed light on the reality of existing solutions, we ran a test protocol with real data on the latest solutions from Kofax, Abbyy, Nanonets, and others.
The combination of the latest OCR models, Transformers, and generative AI demonstrates its unmatched performance.
Observed performance in a production environment:
98.9%
Automatic document splitting
99.7%
Automatic document identification
99.8%
Compliance point checking accuracy
98.4%
Extraction of typewritten data
94.2%
Extraction of handwritten data
0.09
Processing seconds per page
02
GDPR and Data Security
The confidentiality of your sensitive data
GDPR, HIPAA, cybersecurity... you can't always trust a cloud solution to guarantee the security of your confidential documents, personal data, or health data.
Our on-premise solution uses a local architecture that ensures the protection of your most sensitive data. All processing operations are performed on your own infrastructure, thus guaranteeing the confidentiality and security of your documents.
Our on-premise solution uses a local architecture that ensures the protection of your most sensitive data. All processing operations are performed on your own infrastructure, thus guaranteeing the confidentiality and security of your documents.
Download our solution (Docker container)
Our solution is freely downloadable, and all its functionalities can be used in your environment. There are no limits in training capabilities, but restrictions apply to volumes beyond an inference of 5,000 documents.
Our solution is freely downloadable, and all its functionalities can be used in your environment. There are no limits in training capabilities, but restrictions apply to volumes beyond an inference of 5,000 documents.
Simplified Integration
Once your AI model is trained and deployed, you can simply call your API from your applications to perform segmentation, verification, classification, or extraction to customize your business workflows.
1
2
3
4
5
6
7
8
9
10
11
12
13
2
3
4
5
6
7
8
9
10
11
12
13
from provenceAI import Model
my_model = Model(
api_key="my_key",
model="my_model")
results = my_model.process(
"my_document.pdf")
print(results)
{
"pages": [
{
"page": 0,
"class": {
"value": "Contract",
"confidence": 0.99 },
"extractions": [
"date": {
"value": date,
"confidence": 0.90 },
"signed: {
"value": true,
"confidence": 0.95 }]
}
]
}
Whether it's automating tedious manual tasks or providing actionable insights, our platform is designed to meet the diverse needs of organizations in various sectors.
Public Services
Automate the processing of documents containing personal and health data to free up staff, better serve citizens and patients.
Financials
Automate the processing of confidential financial documents (consumer, corporate) and conduct anti-fraud checks.
Insurance
Speed up the processing of claims, enrollments and underwriting to streamline customer relations and reduce the burden on underwriters or claims handlers.
Businesses
Simplify the automated processing of HR, accounting, and other massive document flows.
03
· CAPABILITIES
Our platform is used by large organizations that need to process massive volumes of documents for identification, classification, complex data extraction, or fraud prevention.
Features in free version Frioul v.1.2
- Creation of classification, extraction, or combined IDP models.
- Unlimited annotations (document and extraction labels).
- Model training and inference (unlimited trainings, inference limited to 5,000 pages).
- Image correction (orientation, skew, noise...).
- scoring : readability scores for images, confidence scores for classification and data extraction.
- User arbitration interface: you can correct the output according to a confidence limit.
- Performance analysis interface by labels.
- Retrieval of extracted data (.csv).
- splitter : automatic splitting of pieces on trained documents.
- splitterAI : agnostic zero-shot piece splitting (alpha).
- quickLearnAI : similarity classification engine.
- fewShotAI : few annotations, thanks to generative AI applied to annotated documents, to multiply performance.
- quality : module to validate the quality of extracted data, according to criteria for each data type.
- API access.
Additional features in custom version Massilia v.1.1
- rights: advanced user role and rights management module for the platform.
- reTrain: intelligent model retraining planning.
- autoImproveAI: the model automatically self-feeds to improve performance.
- arbitrageAI: the arbitration interface includes advanced features such as model retro-improvement, the possibility to arbitrate files (session of several documents), workflow replay, suggestion of new labels on unrecognized documents, correction on historical documents.
- flows: you build in-app advanced data processing workflows.
- compute: you build calculation rules for data validation.
- lookup: universal connector for querying third-party services (API, SQL).
- truthOrDareAI: document authenticity validation module (image forensics).
- fraudAI: fraud detection models (documentary inconsistency, intra-file inconsistencies, inter-file suspicions) based on business rules.