AHRC implements a fully automated EDRMS
The Commission uses AI and machine learning to classify records without the need for staff input.
Australian Human Rights Commission
The Australian Human Rights Commission is an independent statutory organisation, established by an act of Federal Parliament. They protect and promote human rights in Australia and internationally.
These efficiencies will allow the Commission to deliver better service to the Australian public as the AHRC conducts its human rights compliance, education, and public awareness activities.
Chief Information Officer
AHRC implements a fully automated EDRMS
Implementing intelligent records classification
According to the National Archives of Australia (NAA), the Australian Human Rights Commission (AHRC) is the current thought leader for records management in the Australian Government. Through a corporate partnership with RecordPoint, the Commission implemented an electronic record and document management system (EDRMS) on SharePoint Online with RecordPoint Records365, utilizing RecordPoint’s AI and machine learning technologies to classify records without the need for staff input.
Surveying the landscape
The AHRC is Australia’s national human rights institution, an independent statutory organization in the Attorney-General’s portfolio that promotes and protects human rights. Although a small agency, AHRC has a reputation for early strategic adoption of new technologies. In recent years the Commission has been a leader in government with its migration to Microsoft Office 365 and Azure.
Prior to this new EDRMS solution, in production since February 2019, the Commission was drowning in a sea of duplicates, tangled in nested folders, and perplexed by lost documents. Funding shortfalls and other challenges saw the Commission unable to implement an EDRMS solution that was viable.
Researching options, the Commission, headed by Ron McLay, Chief Information Officer and Ryan McConville, Information Manager, incorporated the Department of Finance’s study into the failings of the traditional EDRMS. In particular, the report suggested that records management should be automated, rather than a being a manual task for public servants. Inspired by the report, the Commission set out to implement a fully automated EDRMS, using artificial intelligence (AI) and machine learning. This would form the basis of RADICAL: Record And Document Innovation & Capture – Artificial Learning.
RADICAL is the first project of its kind in Australia.
An EDRMS implementation with AI addresses key problems faced in running an effective EDRMS. AI reduces the scope for human error, while increasing the volume, accuracy, and consistency of records classification. The simple user interface has driven high user uptake and is seen by staff as a useful tool rather than a burden.
The Commission avoided customization and add-ons for RecordPoint and SharePoint Online, focusing on configuration instead. A common problem agencies have experienced is in the customization and use of third-party add-ons to suit existing or outdated business processes. This often resulted in systems that were difficult to use, inefficient and unreliable, and hard to upgrade, with user uptake suffering accordingly.
Harnessing the native functionality of RecordPoint and SharePoint translated to improved business processes. The Commission also incorporated simple navigation for easy browsing of records, supported by a powerful search feature in Records365. The RADICAL solution has been in production since February 2019 and deployment is being staged across the organization.
The Commission set out to implement a fully automated EDRMS, using artificial intelligence (AI) and machine learning. This would form the basis of RADICAL – Record And Document Innovation & Capture - Artificial Learning.
Outlining a solution
RADICAL will greatly increase the efficiency of the Commission’s work through streamlined and digitized business processes, improved corporate reporting, better retrieval and reuse of information, and ensuring timely, compliant disposal of dated or low-value records. These efficiencies will allow the Commission to deliver better service to the Australian public as the AHRC conducts its human rights compliance, education, and public awareness activities.
The successes seen with the Commission’s project RADICAL also demonstrates to other government agencies that AI/Machine learning is a viable, affordable option for automating records management in the public sector. This is significant, as historically the high costs of implementing AI-based systems has been beyond the reach of most government agencies.
RADICAL also aligns with the key outcomes of the Department of Finance’s Digital Records Transformation Initiative, namely to:
- Make effective use of the smart technology that has emerged, but not yet been incorporated, into records management practices;
- Improve productivity using automation
- Increase the reuse of information assets across Government
- Increase compliance with regulations for the management of Australian Government records
Before RADICAL, the Commission managed its corporate records in paper files and electronic file shares. The paper file was considered the primary file, while electronic copies of those files were kept for ease of reference and sharing.
The process of creating and sentencing paper files was time consuming and relied on staff members with limited experience, and often no interest in records management, to make accurate decisions about the retention and disposal of valuable corporate records.
When planning RADICAL, a key goal was to remove records management decisions from staff and allow them to focus on their core work. RADICAL needed to provide “transparent records management” and limit the potential for inaccurate or inconsistent classification.
Many commercial EDRMS platforms are feature rich but can be difficult to use and these features may not align with actual business needs. Agencies will often compensate for these shortcomings through customization and add-ons.
The Commission’s approach was “configuration over customization” as recommended by the DTA, focused on human-centered design. Staff were consulted extensively on current needs and pain points. When possible, native Records365 and SharePoint functionality was preserved, limiting the need for end-user training and burdensome change management.
Approach to implementation
The project methodology followed the NAA’s recommended approach to implementing an EDRMS. The initiation phase consisted of an exhaustive review of the Commission’s existing information management environment. The review included:
- An assessment of the Commission’s information management policies and procedures
- A review of the Commission’s 2003 agency-specific records disposal authority
- The documentation of gaps in practices in comparison with the NAA’s Digital Continuity 2020 Policy, with the assistance of the Check-Up 2.0 self-assessment tool.
The Commission developed a business case that defined how an EDRMS could:
- Improve record and document management practices
- Comply with the Digital Continuity 2020 Policy (DC2020)
- Engage with stakeholders
- Engage the executive-level staff as major stakeholders
- Create and utilize a steering committee
- Staff the project team with the appropriate skills and knowledge, including an experienced records manager
- Identify and address risk associated with implementing the solution
- Mitigate the risks such as change management/user uptake and technology failure through extensive stakeholder consultation and testing and avoiding customizations and use of third-party products
Previous attempts to launch an EDRMS within the Commission failed for a variety of reasons, including resourcing issues and failure to gain executive support. When planning RADICAL, the Commission secured the sponsorship of the Chief Executive and ensured that a senior executive was assigned to the project.
Taking a human-centered design approach to the project ensured:
- Ease of use really was the overarching principle of the implementation and not a mere project cliché
- No mandatory actions or metadata would be required by users
- No use of third-party products or system customizations
- Records management was automated and transparent to the user, utilizing rules and machine learning
- Users were extensively consulted to ensure that the system met their business needs, improved processes, and enhanced corporate reporting
- Extensive piloting of the new system
Consulting with NAA and the Department of Finance regarding the proposed solution, and with several other agencies to identify their points of pain and lessons learned, the Commission built the system to avoid these pitfalls wherever possible.
The implementation phase involved:
- A detailed analysis of existing systems and record holdings
- Developing a new information governance framework and agency-specific records disposal authority
- Developing and implementing a records migration strategy
- Extensive collaboration and testing with the RecordPoint AI developers
- Development and testing of the SharePoint platform that Records365 manages
- Working with a specialist change management facilitator
- Training end users and providing ongoing support on go-live
- Implementing an agreed security model
- Gaining agency-wide approval for the system including the business rules and use of the machine learning algorithms
- Training the machine-learning algorithms
- Implementing and rolling out the system
Rethinking traditional records management
Records classification involves categorizing records by function and activity as set out in the Administrative Functions Disposal Authority (AFDA Express).
Traditionally, the classification process has been performed manually by records officers. The manual element of classification can be time consuming, can lead to inaccuracy, and can be disruptive to staff. Previous methodologies to automate records classification used rules trees to classify records based on their metadata and saved location. However, rules trees needed to be built and maintained by experienced records officers and relied on end users to apply accurate metadata and save to specific locations.
Leveraging AI in this process solves many of these problems by combining a minimal rules tree with a machine-learning model. If a record cannot be categorized by a rule, the machine-learning model classifies the record based on its contents. This system eliminates the need to maintain complex rules trees, and the reliance on metadata and record location.
The RADICAL project team worked with RecordPoint’s AI developers to create a statistical model that can classify records against AFDA Express and the Commission’s agency-specific records disposal authority.
The statistical model is developed by taking a set of records that have been manually classified and applying natural language processing techniques to normalize the document content into vectors. The model is then trained using algorithms.
After an initial training period, the RADICAL statistical model can categorize individual records with an accuracy of 80%. The Commission expects this accuracy will increase over time. RADICAL also recategorizes records each time they are edited, ensuring the classification is always current.
Although the machine-learning model will initially work in conjunction with a rules tree, as the accuracy of the model increases the rules will be gradually removed and the Commission will rely solely on machine learning to manage its corporate records.
RADICAL provides multiple, tangible benefits to the Commission such as:
- Automated records management that is accurate, consistent, and compliant
- Compliance with DC2020
- Document versioning, which reduces duplication
- Enhanced collaboration and sharing
- Streamlined handling of Freedom of Information (FOI) requests
- Power BI reports for senior executives
- Reduction in staff time spent on records management
- Effective and efficient records search and retrieval
- Real-time video transcription
- Automated image cataloguing
As most Australian Government agencies share the same records management requirements, the Commission feels that the machine learning model provided by RecordPoint and used by RADICAL is a “genuine game-changer” and will allow other agencies to experience equivalent “gains in efficiency, productivity and cost reductions.” The Commission sees itself as a trailblazer in government for the use of AI in records management and is excited to share its experiences as the current thought leader of Australian Government records management.
RADICAL has had a positive impact on the Commission and its stakeholders by:
- Delivering on the objective of “transparent records management”
- Increasing the accuracy and compliance of information management practices by reducing the scope for human error
- Reducing the time and costs associated with responding to FOI requests through improved search and retrieval
- Gradually reducing physical storage costs, currently averaging $17,000AUD per year
- Reducing digital storage costs
- Increasing collaboration between Commission business units through shared document libraries and the establishment of an “open by default” information access policy, where access to records is restricted only to protect personal privacy or sensitive information
- Improving business processes through electronic workflows, document versioning, and automated metadata tagging
- Minimizing the impact of potential data breaches through regularly scheduled records disposal
Initial estimates by the Commission suggest that staff using RADICAL are seeing at least a 5% increase in productivity. Additionally, the accuracy of capture and classification by the algorithms is improving, and by estimates “it already exceeds the accuracy of our manual classification.”
Lastly, the Commission showcases that a technologically advanced solution can be implemented without significant costs. “We estimate that a traditional EDRMS would have cost the Commission 3 or 4 times as much as RADICAL.”
Challenges and lessons learned
The Commission consulted extensively with other agencies on their EDRMS projects, which was invaluable in helping avoid similar mistakes. These consultations influenced the choice of a human-centered design approach and highlighted the importance of not attempting to recreate old business processes in a new system. A “greenfield” implementation was a strategic advantage to the design and change management process. RADICAL presented an opportunity to use an advanced AI-driven platform to deliver an easy, modern, and powerful platform without staff preconceptions and complex data migration.
Compiling a training dataset for the machine-learning model proved to be another challenge. In order to provide a learning dataset for Record365’s learning algorithm, a minimum of 1,000 electronic records was needed, individually classified against each class in AFDA Express and the Commission’s agency specific authority.
As the Commission’s primary files are paper, no electronic records had been previously classified. While a problem at first, this was also an advantage as the Commission could ensure the dataset had been classified accurately and consistently, in turn increasing the quality and accuracy of the model.
Another challenge was that the model accurately classified records according to their content but cannot yet factor in the context in which they are created. For example, legal advice about a procurement process will be classified under the AFDA Procurement function rather than Legal Services, as its contents primarily concern procurement. Additionally, where the model identifies the correct function, identifying the correct disposal class often requires a subjective assessment beyond the capability of the Commission’s current model.
However, in working with the RecordPoint team, the Commission was able to address these scenarios and have several promising options to test.
With the Commission’s full backing of Project RADICAL, a prioritized rollout schedule based on risk assessment and mitigating operational impact is underway.
In conjunction with completing the rollout, the next steps include:
- Testing machine-learning enhancements with RecordPoint including the use of deep-learning algorithms
- Digitizing paper-based processes including Commission meeting papers
- Making greater use of Power BI to visualize and monitor key business information for senior staff
- Reducing storage of paper files, either through digitization or regular sentencing
The Commission is also looking for opportunities to further leverage Microsoft machine-learning technology to improve the accessibility of its records. For example, the Commission has been testing machine-learning services to transcribe audio-visual records to text, and to translate some publications to Easy English.
Already a reference agency for cloud adoption, the Commission looks to share its knowledge and experience from Project RADICAL with other agencies.
View our expanded range of available Connectors, including popular SaaS platforms, such as Salesforce, Workday, Zendesk, SAP, and many more.
The ultimate guide to Personal Data, Personal Information, Personally Identifiable Information and Sensitive information
Organizations need to embed privacy into their systems and processes to gain an advantage and gain customer trust. But first, they need to understand the sensitive data they have, and how to classify it. This means they need to learn to separate their PI from their PII. This guide explains the differences between each of these terms.
ChatGPT: Is this popular new technology a threat to privacy?
In this month's FILED Newsletter, how do AI tools like ChatGPT impact privacy? 5 questions boards and execs should ask about data privacy. Looking for a new dev role? Try the dark web.
Assure your customers their data is safe with you
Protect your customers and your business with the Data Trust Platform.