Attivio Platform 5.5: Machine Learning Relevancy and Insight Rule

Attivio 5.5—the latest release of our market leading Cognitive Search and Insight Platform—has a lot going for it. If you just want a quick summary of all its new features, the launch press release is a good place to start.

But I’m going to focus on one—the platform’s use of machine learning to improve relevancy. After all, relevancy is the heart of cognitive search.

Machine Learning Relevancy in the Attivio Platform

Relevancy Is in the Eye of the Beholder

There are a couple of ways to look at relevancy when it comes to search. There’s relevancy to a topic. And then relevancy to the user. Of course, there’s overlap here. But two users can run the same search query and need quite different results to return at the top of the list.

One of the biggest contributions to profile tuning is user behavior. The platform looks at aggregate user behavior to evaluate how relevancy models are working. User behavior tells us what documents users like, which is important.

But user behavior doesn’t tell us why they liked the documents. To understand why, we combine machine learning with advanced text analytics. That way, the platform can build relevancy models that adapt to changing conditions by incorporating behavioral and contextual cues.

Using machine learning, the Attivio Platform calculates relevancy based on dynamic models for each user. Over time, the platform fine tunes these models making sure each end user gets accurate personalized results that match their behavior.

The Attivio Platform provides tools to support the search architect in managing and maintaining relevancy models and also offers manual relevancy tuning to optimize results. The goal is to increase productivity and simplify the use of this capability.

Webinar and Free Trial Download

I’ve just skimmed the surface of what Attivio 5.5—the Sherlock release—can do in terms of improving search relevancy. Plus, the platform packs a lot of other improvements and new features like rules-based tagging and policy mapping.

The best way to learn more about the platform is our webinar coming up on Thursday, June 15, at 2 p.m. ET. Attivio co-founder and CTO Will Johnson and I will host it. It will include a demo and Q&A.

You can also test this new release for free. We provide sample data sets or you can use your own RSS feeds, file systems, databases, or CSV files. With our wizard-based interface, you can go from download to cognitive search in under five minutes.

By the way, on the same download page you can connect to a sample version of our knowledge management (KM) application.


Tweet This

Attivio Platform 5.5: Machine Learning Relevancy for #CognitiveSearch