Making Common Sense out of Search with Semantics on

  • 272
Making Common Sense out of Search with Semantics on

We all live with keyword search and to varying degrees share common frustrations around our search experiences.

Out of necessity, we have had to grow skilled at filtering our own search results by continually tweaking our keywords, and searching within our searches. When the search engine misses the mark, our computer crashes, or we get distracted by Twitter and forget what we were really looking for, we often repeat the whole process from scratch.

In fact, according to a recent MIT study, almost 50% of people repeat the same searches multiple times a week or month. In this post I’ll aim to represent the end of these frustrations by discussing new technology that can fundamentally change our concept and expectations of search.

I’ve made my thoughts on current search models pretty clear in a previous post here.  But I know it’s not just me that’s concerned about the keyword model of search. At Google’s recent Searchology event, Marissa Mayer, Google’s VP Search Products and User Experience, listed the following as the most difficult unsolved problems today in search:

  • Knowing what you’re looking for and finding the most recent info on it
  • Expressing that you want just one type of result and assessing which results are best
  • Expressing your searches in keywords

My immediate question is this: why are we re-tooling an evolutionary process of asking limitless questions (i.e. search) into a finite process of using ambiguous (keyword) terms? Why don’t we think beyond the keyword entirely, and re-frame the process around retrieving information by asking a question and getting an answer. Better yet…“the answer”, and personalize it to the unique individual who asked it.

With that in mind, I’d like to introduce a project I’m working on that’s launching this summer, a site called The stated function of the site is to “reverse the search” by filtering and focusing everything on the web down to only what matters to you the very most. We’re approaching this with the following process:

  1. Abstract the semantics of language around what you are looking for and enrich this understanding with commonsense reasoning. This is all conducted behind the UI using proprietary technology.
  2. We then apply learning systems that continue to refine the process of determining meaning as it relates to you as an individual – a process that continues after you are on to the next task. With each site visit the system learns from you, incrementally enhancing your results every visit.
  3. We’ve incorporated social technology with our learning systems to connect you to others that may have found the answers to the questions you have, or have accomplished tasks around the things that matter to you.

Example: Jennifer wants to travel to Paris [click to enlarge]

Dorthy example

Keywords might be good for a discovery system, or in a “like model,” but in providing a true answer you need to be able to have a conversation. How we consume information from around the Web is constantly changing, and search must be a partner in that evolution.

As the Chief Technology Officer for, produced by Saber Seven, Jim Anderson leads the company’s search, content delivery, social and mobile strategies, as well as technological vision. The mixture of his experience, with leading advances in Artificial Intelligence, Natural Language Processing, Machine Learning Technologies, and his hundreds of patents have all been fertile training for his new challenge – creating as a premiere web destination, helping users to find real answers to their searches.


Subscribe to SEJ

Get our daily newsletter from SEJ's Founder Loren Baker about the latest news in the industry!

Read the Next Article
Read the Next