Hakia, a Semantic Search that Searches the "Talk"
By Lark Birdsong
Birdsong Research
Overview
A semantic search engine takes the content of a search query and delivers relevant information based upon the overall meaning of the search versus the frequency of a few key words. The query is based upon our natural language and the results hopefully bring back information that is more relevant. Hakia (http://www.hakia.com/) is one such company and wants their technologies and systems to function with human-like intelligence when we put in our search terms. All this assumes that we, as end users of the search engine, put in a complex query for our search when we are using a semantic search engine.
Hakia founded in 2004, is one of several semantic search engines and in 2008 was one of c/net’s Webware top 100 best Web 2.0 applications (http://www.webware.com/html/ww/100/2008/winners.html). Hakia’s technology will query the entire web and Hakia notes that they differ from other search engines in two ways. They use a meaning based search and analyze and store the web page content in terms of ‘knowledge bits” using their proprietary QDEX system. Secondly their “SemanticRank algorithm is comprised of innovative solutions from the disciplines of Ontological Semantics.” Per Hakia, this provides higher quality results than other search engines that don’t use semantic search technology. Hakia has three criteria for quality results. “It (1) comes from credible sources (verticals) recommended by librarians, (2) is the most recent information available, and (3) is absolutely relevant to the query.”
Search Example
I tested Hakia out with some research regarding the early onset of arthritis in women. The power of a semantic search engine lies in its ability to handle more complex searches so keep this in mind when planning your query. Using my example of above, I wanted to compare a couple of medications for people with arthritis and used this query in both Google and Hakia. My search query was to “compare Corticosteroids and Nonsteroidal anti-inflammatory drugs (NSAIDs) for women with arthritis.” In Hakia several relevant articles came up that did compare these different approaches to pain relief for people with arthritis. Google did return more keyword search results and they were not as relevant for the complexity of my query, although a few articles were helpful.
I would have not used the full capabilities of Hakia had I only put in a few of the key words such as arthritis, women, and drugs. So don’t let the search box in Hakia fool you into using only a few key words as you want to test out complex queries. I wish they had something that caused the searcher to think “big or complex” with their query. Maybe a big box (unlike the smaller search box we are used to) that says something like “Your Complex Query Please!” As a side note Hakia also gives you a gallery for what they deem to be the main concept. In my search it was for the word arthritis. They further cataloged the search on arthritis into several categories such as news, symptoms and treatments, clinical trials, alternative treatments and more. I found this to be useful as well. In essence, I ended up with a keyword search along side my more complex query.
Thus we add another tool for us to use with Hakia in the wonderful world of information seeking. Welcome back from summer for those in the Western Hemisphere and I hope everyone had a great time these past few months no matter where they live.
Should you need assistance with information
or research please contact me at
or 303-884-8666
