2 min read time

Putting Machine Learning to Work for Social Media Governance

by   in Data Analytics

social-media-2537391_640.jpg

The prevalence and widespread use of social media continue to grow. From a business perspective, social media provides a platform to quickly and efficiently share information to a wide breadth of different audiences. 

Unlike traditional data sources, social media is often made up of unstructured data that can incorporate text, video, audio and images. The explosive growth of this increasingly diverse new data is taxing corporate resources and putting enormous pressure on the compliance office. Social media, regardless of its form, is held to the same regulatory compliance and discovery demands as traditional corporate data and, as such, needs to be monitored and managed with strict information governance policies. As a matter of fact, in an article published on the American Bar Association website, Joseph F. Marinelli, a partner at Fitch, Even, Tabin & Flannery, writes, “The Notes also make specific reference to social media, suggesting that counsel should be familiar with their clients’ information systems and digital data – including social media – to address proportionality and preservation.” 

The regulation of social media is prevalent and enforced across a multitude of different industries today. For instance, the Financial Industry Regulatory Authority requires registered brokers­ and dealers to have policies and procedures to monitor electronic communications. In addition, the U.S. Securities and Exchange Commission regulates disclosures made by its members on social media. These are simply a couple of examples of how specific industries are demanding that organizations regulate social media content, which is essentially another enterprise data source covered by regulatory mandates and electronic discovery. 

Because the data is so prodigious, a number of challenges are created for organizations as they look to implement both policies and technology for this paradigm shift. From an information governance policy perspective, organizations need to understand the data, where it is being disseminated (internally versus externally) and how it is being managed throughout its lifecycle. This requires that organizations address social media as a part of their overall governance strategy to ensure that their data policies are properly managed and maintained in accordance with regulatory demands of their respective industries while operating under tight budgetary constraints. 

From a technology perspective, the monitoring of social media is challenging traditional information governance solutions and approaches. New technology is being applied to standard practices and having a significant impact on organizations’ ability to mitigate risk, streamline eDiscovery and ensure a more compliant organization. 

Because social media is growing at such a fast rate and its content is so diverse, the traditionally labor-intensive approach for addressing compliance and discovery requirements is extremely costly and error-prone. This has led organizations to evaluate and implement big data-related technology such as machine learning, which enables computer systems to continuously learn from data and use such learning to uncover patterns and relationships. Organizations can then leverage this capability to automate the classification of massive amounts of diverse data and/or tracking of how sensitive topics may trend across multiple sources. 

By accelerating an organization’s ability to understand what’s in their social media data, machine learning is changing how organizations mitigate the risks inherent in social media, and is providing a way to effectively and efficiently apply information governance policies consistently across all their data. From enterprise content management and compliance to archiving and eDiscovery, machine learning provides the intelligence needed to make the right decisions fast.

Labels:

Information Governance