Haystac: Unified Information Governance for a Sustainable Content Intelligence Platform

Riding motorcycles is an activity that requires the right ‘tool’ (i.e. motorcycle) for the type activity (i.e. ride) selected. “The interests of people who ride motorcycles vary. Some enjoy riding off road, others enjoy racing, and many enjoy the “adventure” riding as a group or building and customizing classic motorcycle models. This is analogous to the world of Big Data. With the variety of business objectives in Big Data, types of motorcycles for different purposes represent the different tools that are used to classify, categorize and analyze unstructured portion of Big Data,” explains Barak Tsivkin, President and CEO, Haystac. Tsivkin is a motorcycle enthusiast and a successful entrepreneur who is focused on the Big Data challenges.

Headquartered in Newton, MA, Haystac aims to solve challenges faced by many enterprises in information governance: data compliance, retention and analysis of unstructured data that is either electronically stored or scanned and captured. Haystac’s technology is based on extensive academic research. It is based on machine learning, advanced, hierarchical classification methodology and robust search engine designed for unstructured data, all on a single unified platform. “The two technologies that most influenced the direction of our platform’s development were based on the need to analyze huge amounts of unstructured data, Haystac incorporated both statistical, text based methods for content analytics and to increase accuracy in image processing and visual similarity clustering using the latest scientific developments infacial recognition techniques,” adds Tsivkin.

Unlike other competitors in the market who focus exclusively on ‘unsupervised classification’ (aka Clustering), a methodology that clusters documents based on their visual similarity, Haystac provides an advanced form of classification that is Genre-Based: incorporating both typographical and textual analysis in addition to visual similarity and text based multi root, deep hierarchy document classification.

The technology that most influenced the direction of our platform’s development was based on the need to analyze big unstructured data

What makes Haystac unique in its offerings is that it provides a comprehensive platform capable of true information governance, and a truly multi-dimensional classification based on building deep hierarchical structure. Haystac’s unified governance platform, UIG360°, has BI solution architecture specifically designed to address enterprise needs for analyzing vast amounts of unstructured, text-based Big Data. It does so by bringing structure to unstructured data with an advanced ‘machine learning technology’ that classifies the data based on user-defined business rules and multipurpose hierarchical structure.

“Not only do we have all the functionality our competitors have, but we bring additional unique capabilities to address the entire spectrum of IG for Big Data,” says Tsivkin. In an interesting case study, it was observed that Haystac’s services helped a financial services company to improve fraud detection; by comparing millions of individual transactions against seven year’s worth of customer data without overwhelming the data warehouse or incurring prohibitive expense. Moving forward, Haystac would strive to stay ahead of the competitors in the IGspace. The company collaborates with leading universities and research centers in the areas of text and image analytics. From a business perspective, Haystac is focused on markets where they will be offering specific IG solution and expanding the company’s partner network. The company provides vertical solutions for energy, HC, financial services and government.


Newton, MA

Focused on the practical application of machine learning and information extraction technologies to solve challenges faced by enterprises.