Discovery Engine

Discover patterns and anomalies in data that take your business to the next level.

simMachines provides a state-of-the-art similarity engine that outperforms other approaches by a large margin in terms of speed and precision. Our technology is the result of more than 8 years of research and development.

In the following experiment, our engine (R-01) is compared to the permutation index (Perm). We created a database of up to 120 million strings of 20 characters each and we used the hamming distance, measuring the time it takes our index to answer a query as the database size increases. The result speaks for itself:


Discovery Engine Experiments
Clustering is useful to understand the real expression of data and we provide a very flexible clustering engine that works on any kind of unstructured data (real values, textual elements, categorical values).
In the visualization above, every slice represents a cluster and every box in each slice represents a topic, some piece of information highly correlated inside the cluster. Boxes close to the center represent topics that are very frequent and boxes close to the perimeter of the circle are less frequent. In this way, we can show many clusters at the same time, and we can also group clusters that share common topics in a hierarchy.
We believe this is very useful for identifying patterns in complex data-sets. This rendering is showing 100 million transactions and our clustering engine completed the result in 2 hours (one machine).

Our engine is capable of answering queries at very high speeds in large databases. Furthermore, query time remains virtually constant. For more information, please read a white-paper that benchmarks our engine against top similarity search data structures.

Features:

  • Fastest query time in the market.
  • Simple to use.
  • Zero configuration & maintenance.

Download Performance White-paper

Your Name (required)

Your Email (required)

Please leave this field empty.

Your Phone (optional)

Network Security & Pattern Inference

Network Security & Pattern Inference

We inferred patterns from large number of strings in order to identify potential Internet security issues.

Auditing

Auditing

We audit complex documents and processes consistently and efficiently.

adminWe Discover