Perhaps the biggest problem with natural language search is that it’s incredibly difficult to try and automate machine-assigned ontologies. Essentially, machines just don’t get it.
(Okay, okay, we just wanted to see if we could sneak the word “ontology” into two posts today.)
Semanti’s founder, technologist Bruce Johnson, agreed: “Machines don’t really deconstruct language well. They miss so many of the ambiguities and they often don’t pick up on synonyms.” (Yet another reason why machine translation is usually pretty crappy.)
So what does Semanti do to combat this? They bring in more humans:
Most semantic search services are natural language search engines; however, Semanti employs a system of personal bookmarks, a drop-down menu with multiple definitions, and search recommendations pulled from Facebook friends. Semanti actually increases relevancy by introducing human eyes and opinions into the search process.
Rather than focus on creating and promoting yet another search engine website, Semanti has been launched as a browser plugin for Firefox that integrates with Google, Yahoo or Bing. It replaces the search engine’s suggest feature with suggestions to find the meaning behind your queries. (RWW’s example is [apple]: the brand, the fruit, the Beatles’ label, Gwyneth Paltrow’s baby, etc.)
Next, Semanti turns to pages you’ve bookmarked with the plugin—pages you’re likely to turn to. Semanti searches the text of those pages for anything relevant. And it learns— the more you search and click, the more relevant its results will be.
To take it to the next level (and the big, exciting thing you’re all waiting for), you can also use Semanti Social Search which connects with your Facebook friends’ Semanti bookmarks. You can turn this off later if this doesn’t end up being helpful.
Read Write Web has a video of a Semanti demo
. . . but apparently I’m not one of the cool kids. (You can try it for yourself, if you want. I’m going over here to sulk.) (It says it’s a private video and you have to be friends with the creator to view it.):
What do you think—is this harnessing the wisdom of crowds and putting it to work for you, or will it end up being too much noise to signal?