If you've been following this thread of posts on entity resolution, you'll recall that we have differentiated between the full integration of entity resolution within a master data management system from the other (perhaps operational) uses that do not require a master data index. While the examples we've looked at are largely drawn from existing, traditional uses of entity resolution, if you want to make use of it as a standalone utility, it makes sense to look for emerging usage opportunities.

Let’s look at three ideas that can augment the value of entity resolution as a free-standing application:

• Scanning text for entities – The current mania around big data often hovers around the expectation that massive unstructured data streams are ripe for analytical exploitation. One requirement, though, is the ability to identify entities that are referred to within these text streams. The implication is that entity resolution must be incorporated into test stream analytics to pluck out entity representations from an unstructured context in near real-time.
• Integration with alternate identifying attributes – The number of aliases and contact mechanisms has exploded since the earlier days of entity resolution. Between the increased number of home, work and junk email addresses; social media handles; IP addresses and numerous telephone numbers, there is a much wider variety of identifying attributes available today that can be folded into the matching process. Even more interesting is the incorporation of other sources of identification, such as vocal tonality, use of vocabulary and speech patterns from streamed audio, or image recognition from pictures or from video streams.
• Integration with behavior patterns – Once you are willing to consider alternative identifying attributes, why not extend the pool of those characteristics to include mined behavior characteristics as an integral part of an individual’s identity? Much like what I suggested about use of vocabulary as an identifying attribute, the same could be said for any recognizable pattern.

The consolidation in the MDM industry has led to the absorption of many entity resolution tools into the MDM portfolio. But perhaps that has created a vacuum in the market for standalone entity resolution tools. Anticipate a new generation of technologies that are adapting to new data realities to demonstrate exciting uses of entity resolution.