Datactics InsurTech - International Feasibility Study

组织必须能够获得高质量、完整、最新和可查询的数据，以执行风险管理、欺诈检测和资源分配等任务。数据正在快速生成，公司通常需要将英国公司之家的开放数据等源与Dun和Bradshaw等第三方数据集成到内部存储库，以便进行尽职调查和KYC等流程。然而，在格式、拼写错误、转置和丢失数据方面，并非所有数据都是平等创建的，这使得集成这些数据成为一项非常重要且耗时的任务。目前的方法通常是在内部临时处理，并生成脚本。这些都是非系统化的方法，导致脆弱的管道难以复制和维护。本可行性研究项目旨在了解保险部门的数据挑战，并在该领域建立网络。一个关键的结果是解决这些数据质量和大规模匹配的挑战，这些挑战与从多个来源以不同格式连接大量杂乱、不完整的数据有关。

It is essential for organisations to have access to quality, complete, up-to-date and queryable data for tasks such as risk management, fraud detection and resource allocation. Data is being generated at a speed and often companies need to integrate sources such as open data in UK Companies House with third party data such as Dun and Bradshaw to internal repositories for processes such as due diligence and KYC. However, not all data is created equally with issues in terms of format, mis-spellings, transposed and missing data making integration of these data a non-trivial and time consuming task.Current approaches are often ad-hoc handled in house with scripts generated. These are non-systematic approaches resulting in brittle pipelines that are difficult to reproduce and maintain.This feasibility study project aims to understand the data challenges in the insurance section and build networks in this area. A key outcome is to addresses these data quality and matching at scale challenges associated with joining large amounts of messy, incomplete data in varying formats, from a multiple sources.