We have all experienced it. You are browsing a retail website but you don’t purchase anything. Later, the same day or after a week, you are on Facebook or another non-retail website and the ads that pop up in your feed or on the side are the exact items you were pondering to purchase. No, big brother isn’t watching. Well, maybe he is, if big brother is the result of predictive analytics. ZDNet brought this news to us in their article, “Mix and match analytics: data, metadata, and machine learning for the win.”
YouTube recommendations are a perfect example of applying advanced analytics on a massive scale to improve a service instead of just encouraging retail therapy. This approach is based on a technique called locality preserving hashing (LPH). LPH is about finding and using hash functions that preserve locality, meaning that they map similar input to similar output. This analysis leverages inherent features of the data and is a valid method for building this type of solution.
Ontologies express semantics in a precise way and can be used to state relationships to increase the predictability of this semantic search-like technology.
Melody K. Smith
Sponsored by Access Innovations, the world leader in thesaurus, ontology, and taxonomy creation and metadata application.
