Our technology is based on many years of research in the field of Information Retrieval, conducted at the Technion – Israel’s renowned technology institute. Our scientific team and advisors continue to push the technological boundaries every day.

What is a Deeptext Search?

Running on the entire work item, rather than keywords, isn’t just a technical change, it’s a totally new technology we named “Deeptext Search”. In short - it means learning what information users need to see at this moment, based on the item they are currently working on, and proactively presenting it to them on screen in real-time.

What's wrong with the search bar?

Keyword search is a technology built for the internet, where it can rely on lots of clickthrough data. In an enterprise environment, where such data is not available, that is just not suitable. This is one of the main reasons that workers spend between 20% to 40% of their time searching for information, and still 49% of enterprise workers report they have trouble finding the information they need!

This problem isn't going away, in fact, with the exponential growth of enterprise data, it's actually getting worse. This is why xFind developed the first fully context-based search engine.

Headings, paragraphs, blockquotes, figures, images, and figure captions can all be styled after a class is added to the rich text element using the "When inside of" nested selector system.

Proactive Knowledge Retrieval is Beyond Search

To obtain the true contextual relevance of a query, a multitude of signals are used in parallel, so that various angles of the information are simultaneously analyzed. 

But a multitude of signals retrieve an order of magnitude more results – some are relevant, and some are not. 

xFind's unique Relevance Engine technology assesses the level of relevance of a list of thousands of search results, indicating the highly relevant information items, and avoids presenting the non-relevant ones.

Is our technology really that much better?

xFind solution achieves much higher levels of retrieval relevance (over 90%) and much faster (within 1 to 3 days from initial integration). 

A test comparing the ability to retrieve the correct tagged answer to a Stack Overflow question, showed xFind achieves nearly 5X the accuracy of major enterprise search solutions, on default settings.

These levels of precision had been obtained and validated in real life customer environments, and on highly complex knowledge sets.

Headings, paragraphs, blockquotes, figures, images, and figure captions can all be styled after a class is added to the rich text element using the "When inside of" nested selector system.

Proven Technology

Cutting-edge Information Retrieval and Machine Learning algorithms developed by Technion researchers

Multi Signal retrieval – analyzes various angles of the information, for true contextual relevance

Relevance engine – indicates highly relevant information, and avoids presenting non-relevant information

Up to 90% precision,  4.8x the accuracy of similar solutions, on complex information as well