Eye candy ;-)

Take a look at a new video we put together to showcase some of the research we do as part of improving SenseArray's recommendations.

What is it?

SenseArray is software that quickly determines the interests of an individual user of a website or service. It then predicts what that individual is most likely to find interesting or desirable. This technique is commonly known as "collaborative filtering". SenseArray is designed to be easily integrated into an existing website or service through a simple but flexible "REST" API.

Why use it?

The success of your business depends on your ability to anticipate what your customers want or need, and provide it to them. When you have a small number of customers, it can be easier to determine what they want, but what if you have a website and your users number in the thousands or millions?

SenseArray allows you to personally tailor the content you present to each of your customers, even if there are millions of them. For a content provider this means more page views, and a better user experience. For a retailer this means significantly higher conversion rates.

What can it be used for?

SenseArray can be used in a wide variety of contexts, including but not limited to:

• Music and movie recommendations
• Targeted advertising
• Matching people in a dating website
• Recommending interesting and relevant news

How does SenseArray differ from other solutions?

Effective for new users, not just existing users

Most collaborative filters require a large amount of data about a user before they can recommend effectively. Consequently they simply don’t work for new users, the very users you need to attract and keep to grow your business.

SenseArray incorporates a novel “tagging” mechanism so that it can use metadata about items or users to provide tailored recommendations, even for completely new users.

This means that SenseArray can consider information like:

• The user's geographic location
• Their choice of operating system and web browser
• Referring websites and/or search engine keywords

Or any other "contextual" information you might have about your users.

Of course, for existing users SenseArray will use all available behavioral information about those users, like what they viewed and what they clicked on, to rapidly refine its understanding of a user's needs and desires.

Powerful well-tested core-engine

Many collaborative filters just don’t work very well. The reason? Building collaborative filters is hard, it requires a sophisticated understanding of machine learning techniques, and a lot of trial and error.

SenseArray’s technology was designed by Ian Clarke, a pioneering computer scientist, developing such technologies as Freenet, the first decentralized peer-to-peer network, and Revver, the first video sharing network to share revenue with video creators. Clarke has over 3 years of experience building collaborative filters.

SenseArray’s core recommendation algorithm has been extensively tested against standardized metrics. For example, when tested using 100 million ratings of half a million Netflix users, SenseArray performed 5% better than Netflix’ own recommendation algorithm.

Easy integration through REST API and JSON

SenseArray is designed for painless integration, employing a REST-style HTTP API, making it easy to integrate with SenseArray regardless of what development tools you use. All data is encoded in JSON, a simple human-readable encoding format for which libraries exist for all major languages.

Your data stays on your servers

SenseArray is stand-alone software, not a hosted service. This means that SenseArray is very low-latency, no need to wait for an over-the-Internet API call. It also has the security advantage that your data remains on your servers, under your control. SenseArray runs on Windows, Linux, MacOS, Solaris, and any other operating system that supports Java 1.6.

Note: If for some reason you would prefer to use SenseArray as a hosted service, please contact us to discuss it.

Low cost with simple pricing structure, and no lock-in

Unlike other recommendation technologies, SenseArray has a simple flat per-deployment pricing structure, including a setup fee, and a monthly fee which includes upgrades, consultation, and support. Additionally, we don't lock you in, if SenseArray isn't working for you, you're only committed up to the end of the month, making SenseArray extremely low risk.

Excellent support and consultation

SenseArray's pricing structure includes support and advice straight from the engineers that built the SenseArray technology. We can recommend the best way for you to integrate your website or service with SenseArray, and provide guidance going forward.

Get in touch!

Email sales@sensearray.com, and ask us about an initial consultation to see how SenseArray can help you. Take a look at the Technology page for more information about SenseArray's technical characteristics and requirements.