Have you ever seen an online ad appear for “Discount Airline Fares” while reading an article about a plane crash? How about an ad for scuba lessons served up next to a news account of a shark attack? It’s because ad placement software works solely on keyword recognition. Or that’s the way it has worked until now.

There’s a new online ad technology in town that can actually tell the difference between good and bad contextual matches.

It’s called Shinola, and yes, it was named by Igor.

From the company’s website:

Shinola is a contextual ad placement solution that automatically matches highly relevant advertisements (product descriptions) to web content generated from Trainable Semantic Vectors (TSVs). Trainable Semantic Vectors are a major breakthrough in representing and analyzing textual information.

Until now, reliable semantic analysis always came at the expense of scalability. Shinola is the first and only product placement technology that is accurate enough to capture semantic meaning, fast enough to power real-time applications, automated enough to maintain in dynamic environments, and scalable enough to handle web-sized problems.

TextWise’s proprietary technology coupled with its intellectual property protection gives us a uniquely powerful capability to enable the next generation of contextual ad placement.

Shinola offers a unique combination of advantages over other approaches:

* Understands the meaning of webpages. It expresses the world by combining over 2000 semantic dimensions that are directly derived from Internet taxonomies.

* Provides a unique signature for every webpage and every advertisement. E-retailers are no longer trapped with 500 other competitor ads that look exactly the same to the placement engine just because they have the same keyword or are dumped into the same category.

* Highlights the topic strengths and subtle interactions of meaning within each document that make that document unique. Since Shinola identifies both focused topic areas and rich sub-contexts, it can find the best match between a webpage and an ad.

* Handles both keyword lists and full text. It can immediately import an existing database of ad keywords and calculate new signatures to get accurate matches against incoming webpages. But it can also leverage additional text (marketing collateral, ad copy, product literature, customer descriptions, etc) to create even better experience with Shinola.

* Bypasses the computational complexity of deep natural language understanding. Full natural language processing is practical only for relatively small amounts of text, not for real-time systems that need to handle millions of webpages a day. Shinola uses just the right amount of scalable semantics and linguistic knowledge to capture the meaning and content of documents.

* Avoids the classification failures associated with categories that do not align with customer needs and ontologies that rapidly become obsolete. Although Shinola’s semantic dimensions are derived from a broad Internet taxonomy, we do not actually classify webpages or ads into that taxonomy. Because signatures are formed by combining the strengths of hundreds of topic areas, they are immune to minor changes in the underlying taxonomy.

* Overcomes the failure of keyword matching and classification. The failure of contextual advertising to date can be traced directly to the inability of keyword matching and classification techniques to deliver highly accurate, highly focused ad placement. Shinola can find the most appropriate ad for a webpage even if the two use completely different vocabulary.

Quite simple, really. [Read Igor's Shinola case study for more.]