BetaThis is a new service. Your feedback will help us improve it. Tell us what you think

Objective 3Sixty

Objective 3Sixty is a platform to connect multiple information repositories, leveraging the latest technology, to bring consistency, control and searchability to content anywhere.

Keywords

  • Tags: Records management
  • Tags: Automation
  • Tags: Searchability
  • Tags: Efficiency
  • Stage: Development
  • Type: Large Language Model
  • Sector: Public administration
  • Language: English

How does our product work?

Here you find a high-level overview of our algorithmic product and how it works. This overview helps you understand the problem the product is addressing, the intended purpose and the impacts of using it. We also describe the overall logic of the technical system at a high level.

The platform indexes documents where they already live, rather than moving them. It reads each document, proposes a records-management classification and a retention period, and routes anything it is unsure about to an information officer for a decision.

Overview

Objective 3Sixty will allow the directorate to discover, organise and federate information held across more than forty separate stores, therefore managing and extending organisational governance.

The intended benefit is that a freedom of information request can be answered from one search rather than forty, and that records reach their statutory disposal date without a person having to remember them.

Owner and responsibility

Public Records Directorate

Information and Records Management

Records Governance Team

Christopher Crowther

records@example.gov

0300 000 0000

See suppliers and auditors

Suppliers

Objective Corporation

Northbridge Integration Partners

Auditors

Internal Audit Directorate

See more information

More detailed information on the system

Here you can get acquainted with the information used by the system, the operating logic, and its governance in the areas that interest you.

System description

What is a records classification?

A classification decides how long a record is kept and who may see it. Getting it wrong risks either premature destruction or indefinite retention, so proposals below the confidence threshold always reach a person.

The system combines a search index with a classification model. The index is built by crawling connected repositories on a nightly schedule. The classification model proposes one of 214 record types from the retention schedule, together with a confidence score.

Proposals below the confidence threshold are never applied automatically. They are queued for an information officer, whose decision is written back as training data for the next model revision.

References

Public Records Act retention schedule, 2024 revision.

Internal model card and evaluation report, available on request from the owner named above.

Was this information useful?

Would you like to give feedback? Your feedback will help us develop our algorithms further.

If you wish for the system owner to get back to you personally, please contact them directly using the email address above. For more general comments, use the box below.