L&D Key Features:

Internal and External Data Integration and Analysis

Sophisticated non supervised data cleansing and mapping that consolidates relevant claims and policy data using data quality routines

Daily retrieval of external data and extraction of unstructured data using OCR and NLP

Identify deduplication and reconstruction of family households

Social network construction (graph clustering)

Claims / policyholder and households history comparison

Group history claims comparison for employee and commercial schemes

Insurance trained AI/ML models and sophisticated predictive analytics

Comprehensive library of scenarios covering loss of earnings, sick leave, critical illness, life, death and funeral products

Supports individual (retail), small and large employee schemes (commercial)

Analysis on claims payable, in progress or recently closed

Automatic anomaly detection and natural language processing of text variables.

Continuous-learning AI/ML models designed to uncover complex fraud and increase accuracy of alerts at the prepayment and post-payment level

Network and collusion discovery and analysis

Network link analysis to uncover unknown linkages or relationships among seemingly unrelated claims, providers and members

Interactive network visualisation allowing investigators to explore network nodes interrogate relationships and add notes

Investigator dashboard with detection results, alerts and full context

Prioritised alerts with clear scoring and supporting alert and investigative information

Configurable investigator dashboard detailing key information regarding the alert including the projected value

In depth alerts detail, showing all attributes and characteristics that have led to the alert score, links to external and internal information found to be relevant and recommendations of next investigative steps.

Reporting

Custom data exploration and reporting including dashboards showing alert detection, qualification and conversion rates, team workload, trending and scenario performance. Confgurable filtering and display.

Case management and workflow

Integrated case-management features for task assignment, team management, communication and collaboration

Document and correspondence management for requesting and storing key documentation and instructing internal teams.

Rapid deployment and dedicated team of Shift experts and DS

Options to run scenarios in batch or real time, real time alerts investigators prior to claim settlement allowing payments to be blocked if necessary.

4 months average implementation time frame with rapid time to value and ROI

Continuous scenario performance monitoring with dedicated Shift DS and CSM teams and where needed scenarios are further fitted to client data to further improve detection and recovery success.

Using a SaaS model, the Customer Success Manager, the Shift Data Scientist, and Shift claim fraud detection’s ML algorithms continually update fraud scenarios to stay current with evolving fraud trends and emerging schemes