Entity Resolution

Bob R. Smith II & Robert Smith Jr. represent the same person despite the differences in their data.

Entity resolution is the process of determining when real-world entities are the same, despite differences in how they are described or inconsistencies in how data was entered.

The image to the right shows an example of the type of problem that entity resolution helps to solve. The data for these records has many differences, but a sophisticated Entity Resolution system can determine that it is actually the same entity.

State-of-the-art entity resolution also identifies relationships between entities

But entity resolution isn’t just beneficial for finding undiscovered matches. Entity resolution also determines when real-world entities are different despite similarities in how they are described. In this example, the third record only has a one-letter difference from the second record – “Jr” versus “Sr”. However, this one letter is significant, and provides evidence that these are different people.

This happens for a variety of reasons, for example juniors and seniors, twins, or families with Patricks and Patricias under the same roof.  Entity resolution systems are designed to handle these data matching edge cases. 

Today’s state-of-the-art entity resolution solutions use relationships to better improve entity resolution accuracy. For example, Figure 3 below shows a derived relationship identified between Entity 1 (“Jr”) and Entity 2 (“Sr”). Note that not all entity resolution systems would be capable of recognizing this relationship.

Entity resolution determines Robert Smith Sr. (Entity 2) is a related but different entity.

As relationships are detected between the resolved entities, a network of connected entities is developed. This “entity-resolved graph” helps inform future entity resolution decisions. For example, George Foreman (the famous boxer) named all of his boys George. When entity resolution becomes aware of this, matching new George Foreman records can be done with more care. 

Entity resolution is crucial because it solves many of today’s pressing data quality and analytics problems. Utilizing modern entity resolution, organizations can now accurately identify and link entities within and across multiple data sources, despite incomplete and inconsistent data. Entity resolution can also detect relationships between resolved entities. This results in better customer service, more effective fraud detection, greater efficiency, and leaps in competitive advantage.

Entity resolution tackles pivotal business questions, such as:

• Is this new customer application simply the same bad guy we got rid of yesterday?
• Is our supplier actually owned by a globally recognized and sanctioned entity?
• Are these three customers each with one transaction, or is this one customer with three transactions?

Entity resolution is crucial in enhancing the quality and utility of data related to individuals and organizations. By employing techniques like fuzzy matching, data matching and Entity Centric Learning^™, entity resolution offers numerous benefits for organizations in today’s data-driven world.

• Enhances Data Quality and Accuracy
• Creates 360-degree Views of Customers and Other Entities
• Improves Decision-Making and Builds Trust in Data
• Bolsters Analytics for Discovery and Insight
• Enriches Graph Analytics and Visualization

The value and benefits of entity resolution are multifaceted and substantial. From enhancing data quality to improving decision-making and analytics, entity resolution is a key enabler in harnessing the full potential of data in the modern business landscape.

Ignoring or not properly implementing entity resolution can lead to significant consequences for organizations. Here’s an exploration of some of the potential pitfalls and dangers:

These risks underscore the importance of investing in robust and accurate entity resolution systems. These systems aren’t important just because they improve business efficiency and customer satisfaction; they are also critical for maintaining security, compliance and trust.

These use cases demonstrate the importance of entity resolution in various business processes, highlighting entity resolution’s role in risk and fraud detection, customer 360, graph analytics compliance, and more.

Risk and Fraud Detection

Entity resolution plays a critical role in risk and fraud detection by figuring out who is who, and who is related to who – often despite international obfuscation. With entity resolution, organizations can proactively pinpoint bad actors and mitigate risks before they escalate. For instance, in financial institutions, entity resolution matches and links disparate pieces of information to help uncover fraud rings or insider trading activities.

Customer 360 & Marketing Analytics

Entity resolution is critical to delivering customer 360 solutions that enable more effective customer experience, marketing and other programs. Entity resolution unifies disparate customer identities to form a coherent, accurate 360-degree view of each customer. These solutions are crucial for efficient customer engagement across various touchpoints, ensuring that each interaction with a customer is informed by a complete understanding of their history, preferences and needs, which leads to better service and customer satisfaction.

Graph Analytics

Combining entity resolution with graph technologies makes complex data networks more comprehensible and actionable. This synergy enhances the clarity of graphs for both humans and machines, leading to more accurate and insightful graph analytics and graph visualizations. Entity resolution also improves the performance and accuracy of machine learning models by providing a richer context for data relationships. In practice, this means more effectively identifying trends, risks and opportunities in large data sets.

Financial and Regulatory Compliance

Ensuring compliance with financial and regulatory standards is a complex task requiring precise data matching. Today’s advanced entity resolution systems can provide the data accuracy and consistency necessary to meet government and industry requirements. They help correctly identify customers and their transactions, ensuring financial institutions can adhere to regulations such as the Bank Secrecy Act (BSA), which require KYC (Know Your Customer) and AML (Anti-Money Laundering) compliance.

Master Data Management (MDM), Customer Relationship Management (CRM), Customer Data Management (CDM) and Other Data Management Software

In the realms of MDM, CRM, CDM and other data management software systems, entity resolution can improve the data quality of historical data and eliminate the entry of new duplicate data in real time. Ultimately, with well-resolved entity data in these core enterprise systems, originations deliver higher-quality outcomes at a lower operational cost.

Watch the video below for clear, step-by-step examples of how records about people are matched, identified as related or determined not to match. You’ll come away with an understanding of the intricacies of entity resolution and the power of Entity Centric Learning.

With the rapidly expanding volumes of data, the lack of consistency in this data, and an ever-widening range of use cases, traditional approaches to entity resolution (e.g., batch processing and rule-based tuning) are no longer sufficient.

Here are the top 5 things to get right when shopping for your next commercial-grade entity resolution engine:

1. Simple to Use. Overly complex and inelegant products that require outside expertise or in-depth training to bring on new team members result in project overruns and much higher total cost of ownership. The greater the complexity, the longer it will take to integrate new data sources, both initially and for ongoing operations.
2. Accuracy. This is one of the harder things to test for. Small record tests often misrepresent what will happen at scale. Synthetic data rarely, if ever, produces an accurate metric to assess the true ultimate accuracy of a production system. Best practice is to use real data for accuracy testing. If the volume is over 1M records, consider using a vertical slice (e.g., one geography, everyone with a last name starting with the letter “A”).
3. Minimal Moving Parts. Complicated stacks with lots of moving parts have many failure points, come with more security risks, and are much more involved to upgrade.
4. Real Time. Batch-only entity resolution systems require periodic reloads to integrate new data or removed data (e.g., GDPR deletes).
5. Explainable. Business units making decisions based on entity-resolved data must be able to explain why records matched or, conversely, why they did not. This is not only to enhance their confidence in the system, explainability may very well be required if a business unit is called before a regulator.

Advanced entity resolution technology delivered via an API is easier and quicker to deploy, use, manage and maintain compared to traditional full stack solutions. API-based entity resolution solutions substantially reduce the learning curve to get started, so you can be up and running in minutes and deploy in days or weeks. Advanced entity resolution will also use real time Entity Centric Learning^™, machine learned models and relationship awareness to deliver the best accuracy, time-to-value, and total cost of ownership (TCO) of today’s commercial entity resolution offerings.

Download the Entity Resolution Buyer’s Guide to learn all about how to evaluate and select an entity resolution solution. This guide will give you the knowledge and tools to make an informed decision on which type of entity resolution solution is right for your needs today and into the future.

Senzing is the first purpose-built real-time AI for entity resolution. Senzing software makes it easy and affordable to add advanced entity resolution capabilities to your enterprise systems and commercial applications. The Senzing API provides highly accurate data matching and relationship detection to improve analytics and decision-making, without requiring entity resolution experts.

• Minimal data preparation is required with Senzing entity resolution.
• No tuning, training or entity resolution experts are needed.
• The Senzing API runs on premises, in the cloud or hybrid.
• No data flows to Senzing, Inc.

If you’d like to know more about Senzing entity resolution:

Consult with an Expert – Schedule a call with a Senzing entity resolution expert to discuss your requirements.

Try it Yourself – There are three easy ways to take Senzing entity resolution for a test drive: a simple desktop evaluation tool (for Windows or Mac) and QuickStarts for Linux and Docker. You can install the software, load data and evaluate results in as little as 15 minutes.

When you see how Senzing Entity Resolution AI works, you’ll understand why we provide the most accurate and up-to-date results (that are continually updated in real-time).