A project unusual
Within 2021 Archeiothiki was asked to deliver a project with the below mandate:
- to massively examine a large volume of digitized documents related to a Non-Performing-Loan portfolio,
- to identify the presence of particular documents embedded within bulk digitized pdfs,
- to recategorize particular document categories in a reliably way,
- to isolate and recategorize specific documents identified in clearly defined digital folders,
- to consolidate information from different sources in one Electronic Content Management Platform (ECM) repository.
The request stemmed from a client with a large digital archive that was not digitized to the level of granularity required, resulting in an overall hazy image of the portfolio.
Archeiothiki designed and implemented:
1. A sanitization of selected portfolio documents with the utilization of Archeiothiki’s in-house Artificial Intelligence Product Suite. Particular focus was given to document indexing and retrieval, allowing access to the information contained in the portfolio, not previously accessible to the end-user with an accuracy that overall exceeded 95%.
2. A tailor-made solution, developed on ELO, one of the world-leading ECM Suites, particularly for the import of all portfolio documents, in a user-friendly fashion, rendering all crucial information easy to retrieve and process.
A glimpse of the project size:
- οver 100.000 accounts,
- οver 2 million files,
- οver 11 million pages.
A glimpse of the project complexity:
- designing and customizing an end-to-end mechanism for automated categorization and PDF splitting,
- οffering metadata-driven, robust views of the document archive,
- migrating more than 2 million documents matching all accounts and relevant data, deriving from diverse sources,
- implementing a new, more granular but succinct categorization of the processed content,
- creating efficient workflows for the receipt and processing of future physical and digital documents.
- a quantitative and deeply qualitative portfolio reformation,
- an increased level of granularity,
- a clear and organized view of the portfolio archive,
- a one-stop, smooth UX through the ECM Suite, with a friendly look and feel and consistent navigation experience,
- improved document retrieval timelines to optimize execution of activities,
- fast, targeted user navigation experience,
- the ability to drive strategy, leveraging the contents of the document archive and preventing any misinformed decisions.
With the use of new technologies, we were able to compress a project that would require more than 3,000 man-days of manual work to 45 man-days of development by our AI engineers.
We express our sincerest thanks to all our partners for giving us the opportunity to involve ourselves in such a uniquely complex project and for their invaluable contribution to its successful implementation.