Subscribe to this blog via RSS
Subscribe to this blog via Email
Don’t you love that Google politely corrects you every time you misspell a word (it’s Massachusetts, not Massatusets [face palm]). And how does it know, when you type “Hannah Montana tickets” into the search field, that you’re not looking for Greyhound tickets to Hannah, Montana? (I’m pretty sure no such place exists).
It’s all thanks to the magic known as Latent Semantic Imaging (or LSI). LSI uses algorithms (or, as NPR put it in today’s Morning Edition, “fancy math”) to recognize and keep track of patterns in users’ search queries. What’s the point? This is what allows Google to understand context (again, Hannah Montana! or Hannah, Montana).
At this point, we should all be able to come up with myriad reasons why keyword stuffing is a bad, bad idea. In addition to keeping your keyword density to a reasonable level, however, it’s also important to consider all of the words surrounding keywords and phrases if you want to create quality content. If those words don’t make sense to the fancy mathematician spider-robots, then your content is going to send up a red flag, because it probably won’t make much sense to us less-sophisticated human readers either.
I bring this up today because, according to NPR, a company called Burning Glass is making use of what appears to be similar technology to streamline the job application process. According to the article, the technology is able to look through entire sentences in a user’s resume, taking into account degree, work experience, extracurricular activities, etc.
Much like Google’s use of LSI, all of this is supposed to give the user more relevant search results. Users of traditional career-search websites might especially appreciate this. Take John L. of Boston, for instance: “Just because I have experience in political/government doesn’t mean I want to be an Army Dentist.” Or, Virginia Su M. of Cambridge: “HAILSTORM of cuss words.”
