RecoSense Infosolutions Pvt Ltd.


RecoSense Infosolutions Pvt Ltd.

Automated Document analyzer for unstructured to structured

RecoSense is an AI-focused venture offering AI solutions - Natural Language Processing and Machine Learning for Process automation, Automated Audits, Compliance Management with applications in Financial Services, Aviation, Manufacturing, Healthcare and Media/Telco etc.

DocuSense-AI platform for document analysis and data centralisation. The platform transforms raw unstructured text to structured metacontext. The raw data can be Reports, Documents, Web Feeds, Sensor Data, Device Logs, API Output, News streams etc. The system automates reading of documents with content extraction, classification, Metadata enrichment-key entity extraction, form field mapping to values, data format validations and correlation with historical data, other documents, entities.

We have built and deployed an AI based Document Intelligence solution for data acquisition and processing from Digital Documents, Scanned Documents, Scanned Images, Compliance management. For data post processing, We have leveraged using our indigenous Knowledge Graph for automated meta tagging and critical event detection. The platform has huge potential to work with Govt Projects for document intelligence as well as in private industries as mentioned above.

The platform is used for

· Automated audit of financial statements and extraction of financial data points,

· Manufacturing process forms analysis and maintenance audit

· Pharma safety report analysis

· Property assessment of mortgage loan process automation

· Automate job card maintenance audit in MRO/ M&E

· Document automation in Education and Legal etc

How a Document Analyser Functions

Here is a step-by-step process of how the document analyser functions: Context Definition – Content parsing, Line-item data points, Code values, technical terms, Analysis, Company/Orgs, Fields Mapping

Step 1 - Upload a document You must analyse a document and integrate it into your workflow.

Step 2 - The automatic analysis of data DocuSense classification uses AI algorithms to categorise multi-page documents to get relevant information pages before extraction. It can extract – Element, Summary, Tables, Meta Tags, Symbols, and Handwritten Documents.

Step 3 - Data validation, reviews and processing Data is validated using AI-driven techniques after it is extracted. The process can be consistently upgraded and improved through the review and processing stage for better performance. The extraction results improve, and DocuSense helps validate rules, character sets and various dictionaries.

Step 4 - Insights & Actions Data is extracted and aggregated in the report or interface to the machine or process. The insights can be - Data points summary, Analytics Dashboards, Dynamic Insights, Action Recommendations, Deviation Report, Anomalies report Once your data is ready, you can further explore the possibility of automatically exporting information to different business needs and workflows as the next step in the DocuSense.