Zemanta suggests that the algorithm for finding recommended content is complicated, but it seems to rely mostly on identifying concepts and proper nouns. "We analyze your post through our proprietary natural language processing and semantic algorithms, and statistically compare its context framework to our preindexed database of other content."
The tool should do a better job at detecting the important keywords from the text and should allow users to provide feedback for the suggested links. For now, it's a pretty cool way to find blog posts related to your articles, but I wouldn't be surprised if this will evolve into a clever blog editor.
