Overview

Want to understand which vessels are transacting with whom, where and when? Read our latest case study here.

About Senzing

Senzing is a U.S.-based provider of an entity resolution API that helps customers match records and detect relationships within their data. Senzing® entity resolution software enables customers to include advanced data matching and relationship discovery capabilities in their applications and services without needing to be an expert in entity resolution. The company was formed in 2016 through an IBM spin-out of a skunkworks project and a small team of entity resolution experts that took core technology and built a product focused on extreme entity resolution accuracy, scalability and ease of use. The goal was to make entity resolution available to anyone, anywhere, by overcoming the deployment and cost challenges of existing technologies. The Senzing entity resolution API provides highly accurate, continuous data matching and relationship detection capabilities within and across data sources. The software scales to handle billions of records, works with batch and streaming data, and can be deployed with a few API calls. The Senzing solution requires minimal data preparation with little or no tuning or training, and deployments can be done on premises or in the cloud without sending data or providing system access to a third party.  Senzing is headquartered in Las Vegas, NV with employees who are some of the world’s foremost experts in Entity Resolution and have more than 300 years of collective entity resolution experience.

Senzing Entity Resolution: Unique Capabilities and Differentiation

Senzing created a purpose-built AI for entity resolution that includes two unique properties. First, the ability to make human-intelligent decisions on extremely small and extremely large data sets, without any pretraining or pretuning. In addition, the software gets smarter over time, as it autonomously learns and adapts in real time, without reloading. The artificial intelligence (AI) built into Senzing software is composed of two tightly coupled classes of algorithms: common sense and real-time machine learning.  

Senzing software comes pre-built with common sense that includes principle-based entity resolution and advanced knowledge. The use of principles is a key reason Senzing software does not need training, tuning or experts to deploy into new domains or to add new data sets, new languages, etc. The principles in Senzing ER are based on expected attribute behaviors. Senzing entity resolution assigns these common-sense behaviors to attributes based on the following three expected behavior settings: frequency, exclusivity, and stability.  In a radical departure from other entity resolution methods, the single set of default principles Senzing software provides automatically work as delivered for a wide range of entity types e.g., people, companies, vessels and planes.

Senzing entity resolution also uses real-time machine learning (ML) to get smarter over time. The real-time algorithms deliver entity-centric learning, anomaly detection, and sequence-neutral processing. Senzing retains history and attribute variations for each entity as it resolves new records against existing entities e.g., learning every name, address and phone variation. Over time, based on the accumulated variations, the software learns nicknames, alternative email addresses, common typographical errors, etc., including intentionally fabricated information. Also, Senzing software actively tracks feature statistics in real time as it resolves and relates entities. By comparing actual statistics to expected feature behaviors, Senzing entity resolution detects anomalies and automatically self-tunes to account for them going forward by either assigning them less value or disregarding them altogether.  Based on what it learns about entities and anomalies, Senzing entity resolution continuously evaluates its earlier assertions to determine if they need to be corrected. Sequence neutrality allows the software to self-correct the past in real time, whether it received record A first then B, or record B first then A. Without sequence neutrality, the error rates of other entity resolution systems increase between the periodic reloads required to bring them up to date. With the sequence-neutral processing in Senzing software, your system is always to up to date, overall error rates decrease over time as new information reverses earlier assertions, and reloading is never required.