BACKGROUND

Resolution of out-of-date data, duplicates, near duplicates & incomplete/missing data (continually) as new data on customers, entities, transactions expands dynamically

KEY FEATURES

• Matches and finds similar records across different data sources
• De-duplicates “similar” records using fuzzy matching
• Harmonizes names & addresses and validates with external sources
• Enriches data with 30+ attributes e.g. URL, parent company, key contacts
• Identifies “best match” pair from different data sources
• Detects & fixes errors via ensemble of proprietary ML & NLP algorithms

BACKGROUND

Data from multiple sources moves through several systems & is transformed, which leads to non-traceability of data’s origins, transformations & errors

KEY FEATURES

• Identifies data lineage via functional and format transformations
• Time-series graphs to show parameters & transformations at each “hop”
• Pre-built library of (500+) transformation rules for various data types
• Proprietary ML & NLP algorithms to “reverse engineer” business rules
• Generates exceptions and alerts on incremental data

BACKGROUND

• Financial & operating documents (e.g., balance sheets, income statements, rent rolls, fund & tax statements) are often available in PDF, JPEG or paper formats
• Firms manually digitize them, which is expensive to scale & results in errors

KEY FEATURES

• ML, NLP and graph-based solution to “reverse engineer” formulas
• 40+ pre-trained proprietary algorithms to digitize, validate values & fix errors
• Reinforcement learning on “its own” and via a “human loop”
• Pre-built financial ontology & business rules
• Output maps to required financial templates
• Points out specific areas where the analyst should review

BACKGROUND

Agreements in PDF or non machine-readable format require manual processing to detect non-compliance, which is expensive & difficult to scale; lack of compliance in agreements expose organizations to various risks.

KEY FEATURES

• ML, NLP & graph based culling of attributes from agreements
• 45+ Pre-trained modules to digitize, validate attributes & fix errors
• Reinforcement learning on “its own” and via a “human loop”
• Prebuilt ontology for NDAs, SoWs, MSAs; fast creation of other ontologies
• Search & compare documents; find trends & generate alerts

BACKGROUND

• Duplicate payments, non-compliance & overpaid SoWs
• Current reconciliation process is manual, difficult to scale & leads to errors
• Companies lose average between 0.1% & 0.3% of all payments

KEY FEATURES

• Culling of attributes using ML, NLP & graph based algorithms
• 45+ pre-trained modules to digitize, validate attributes & fix errors
• Determines duplicate & near-duplicate invoices
• Reinforcement learning on “its own” and via a “human loop”
• Prebuilt ontology for NDAs, SoWs, MSAs; fast creation of other ontologies