Our software identifies suspect claims and billing patterns using predictive analytics, robust databases and link analysis in a SaaS model.
Company: Verisk Analytics
Location: South Jordan, UT
Verisk Health's Payment Accuracy Division provides both private and public healthcare payers with a suite of claims editing, fraud prevention and clinical validation solutions. Our real-time analytics platform delivers solutions in sub-second time that will improve auto-adjudication rates, claims processing efficiency, operations insight and cost containment.
Our next-generation prepayment detection software, Nucleus, identifies suspect claims and billing patterns using predictive analytics, robust databases and link analysis in a SaaS model and a unique, productivity driven user-experience. Conducted by medical, coding and investigative experts, our clinical review process individually validates the accuracy of claim recommendations, serving to improve the quality of cases referred to SIUs. Through this comprehensive model of real-time editing, in-depth fraud analytics and clinical review, Verisk Health delivers their customers unequaled savings. Verisk Health is an equal opportunity employer.
- Innovate with guidance from senior members in the development of new capabilities and business opportunities using a multi-pronged approach to fraud and abuse detection including provider, patient, claim and other perspectives.
- Aid in the research of internal and external data sources and work with data services to acquire and store needed data. Execute exploratory analysis of data set including data cleansing, variable creation, variable transformation, etc
- Execute all aspects of model development; including variable selection, non-linearity detection, interaction detection, model creation, diagnostic tests, validation and lift analysis at the task and project level. Design end-to-end experimentation, evaluation, and analysis of various algorithms and models in production as well as under development based on key performance indicators. Exploit multiple predictive modeling techniques and leverage emerging innovations in text mining, network analysis, etc., to augment fraud models.
- Provide continuous stream of actionable analytical information and recommendations on future improvements and identify and use multiple data sources to improve models. Provide pilot testing support and analyze test results if needed. As needed, assist with creation and monitoring of model performance metrics.
- Execute needed business diagnostics to enable productization and deployment.
- Help create content for development partner updates and develop customized models for development partners while maintaining a strong working partnership with domain experts and development partners, as needed.
- Communicate with diverse audiences including technical stakeholders, clients, and possibly at trade industry events.
- Support and train junior team-members.
- Completes all responsibilities as outlined on annual Performance Plan.
- Completes all special projects and other duties as assigned.
- Must be able to perform duties with or without reasonable accommodation.
- Masters degree (PhD preferred) in a quantitative field is required.
- 5+ years' professional experience building predictive models on very large datasets is required. Academic experience at the PhD level could be partially substituted for professional experience.
- 4+ years of experience with statistical analysis and modeling software such as SAS, IBM SPSS, Statistica, or R is required. Experience with datamining platforms such as SAS EnterpriseMiner, IBM SPSS Modeler, Statistica DataMiner, or equivalent is preferred
- Professional experience applying predictive modeling techniques in healthcare or property and casualty verticals preferred.
- Demonstrated experience with data reduction techniques and statistical and predictive modeling methods (e.g., GLM/Regression, Trees, Neural networks, etc.) is required
- Strong verbal and written communication skills are required; must be able to explain technical results to technical and non-technical audiences.
- Demonstrated ability to manage tasks/projects to required deadlines.
- Demonstrated ability to work with diverse business and client stakeholders.
- Ability and willingness to guide and mentor junior candidates.
Apply online here