The goal of this adapt and apply solution is to show actuaries how Dataiku can be used to create an insurance pricing model based on historical claim data, conduct extensive Exploratory Data Analysis, and push their models to production. More details on the specifics of the solution can be found on the knowledge base. This solution makes use of the Generalized Linear Models plugin to enable the training of GLMs in Dataiku’s VisualML interface.
- Easily develop GLM-based risk pricing by designing complete workflows within Dataiku’s no-code VisualML environment.
- Rapidly develop and detailed analytic insights using powerful exploratory data tools, entirely within Dataiku.
- Ensure comprehensive governance and controls with a centralized space for common data sets, tracked changes, and comprehensive model and process review.
- Enable real-time scoring by easily deploying finalized models for use in other systems internally or externally.
- Develop real-time experimentation on model results using the powerful & interactive modeling application
As the insurance competitive landscape intensifies with the entry of new digital native players, the emergence of new risks, and growing consumer volatility, reinforcing efficiency in pricing strategies becomes a must tackle for insurers. Leveraging Generalized Linear Models (GLMs) for consumer claims modeling is a common market practice approach with a deep, rich, and proven track record.
However, historical set-ups for building and approving GLMs are often outdated and lack modern data science and analytic capabilities, acting as a barrier to developing dynamic pricing strategies, embracing specific risks, and opening integration of differentiating data science-supported pricing approaches to attract new ranges of customers.
Retaining the ability to deliver effective and approved GLMs within a modern data science and analytics solution ensures that the value of these techniques is retained, while dramatically improving the efficiency and effectiveness of the teams developing them. This is achieved by enhancing upstream and downstream processes, including data wrangling and API connectivity, improving existing workflows or pricing modules incrementally with machine learning, as and when desired, and establishing effective governance, analytic best-practice, and centralization of workflows, without sacrificing agility. Embarking on this journey acts as the first step to building modern insurance pricing solutions.
- Requires Dataiku v10+
- Prior to installation, your Dataiku instance Admin will need to install plugins. The full list of requirements can be found here.
- This adapt and apply solution can be installed and used right away in one of two ways:
- On your Dataiku instance click + New Project > Industry Solutions > Search for Insurance Claims Modeling
- Download the .zip project file for your Dataiku version and import it directly to your Dataiku instance
- Dataiku Online customers can install the solution from the Launchpad by clicking Features > Add A Feature > Extensions > Insurance Claims Modeling
Work directly to model claims using generalized linear methods inside Dataiku’s no-code VisualML interface.
Use a built-in and user-friendly application to explore in real-time how feature value changes can directly impact model outputs.
Streamlined data exploration
Using a combination of Visual Recipes and Data Visualization, easily process and transform your data to achieve rapid and thorough analytic insight and iteration. Easily integrate large amounts of data from a variety of sources, internally and externally, and explore how they can be leveraged to improve claims modeling.
Easy API Integration
Quickly deploy or update finalized models via APIs for use in other systems internally or externally, including for real-time prediction or response.
Comprehensive metrics allow for simple visual evaluation of multiple models and relative performance.