OpenText has launched an automated classification application to manage the retention and disposition of high-volume, low-touch content such as social media, e-mail, office documents and legacy content.

“When we talk to our customers about auto-classification technology, their biggest misgivings are, 'how do we do it consistently for all content and how do we prove it?' OpenText Auto-Classification has been designed from the ground up to address these problems,” said James Latham, Chief Marketing Officer at OpenText.

“We now have the industry's first machine-assisted classification with built-in statistical sampling and quality assurance to ensure that auto-classification is both transparent and defensible. This solution fundamentally changes how digital records are managed.”

Business and records managers must govern the classification, use, retention, protection, retrieval, and ultimately, the disposal of 'business records'. Increasingly, however, they're being asked to manage massive volumes of 'transitory' or low-value social content and e-mails due to their cost and potential risk.

Classification of content is critical because it lets the business know what content to keep and what can be thrown away. Historically, end-users have been asked to classify content, but adoption and accuracy rates have been low, often leaving the organisation exposed to expensive ediscovery requests and penalties.

OpenText Auto-Classification promises consistent, defensible classification of content without end-user intervention after the system has been set up. It uses the OpenText Content Analytics engine to 'read' through each document, e-mail or social media posting to classify content according to corporate policy and legal requirements. In contrast to search or keyword-based text analytics, OpenText Content Analytics codifies language-specific nuances identified by teams of linguistics experts to dramatically improve accuracy.

It includes workbenches for identifying exemplar documents and rules, testing and refining effectiveness and quality assurance, and sampling against a broader set of documents on an ongoing basis. OpenText says this gives organisations the level of transparency they need to continually adapt the auto-classification engine to their changing needs and to prove to the courts and regulators that the organisation maintains an effective records management programme.

OpenText Auto-Classification works in conjunction with OpenText Records Management so that existing classifications and classified documents can be used in the tuning process.