SBIR Phase II: Applying Latent Group Models to Web Publishing

This Small Business Innovation Research (SBIR) Phase II project will extend the work begun in Phase I to apply advances in knowledge discovery to bridge the gap between what is known about an Internet viewer and what is done with this knowledge to improve user experience and business outcomes. The effort will develop new algorithms to combine implicit and explicit taxonomies to build content networks. A live feedback loop that uses multivariate test results will be used to adjust and refine clusters of users in order to establish specific parameters which can subsequently be acted on. Online content publishers aggregate enormous volumes of data about their viewers from web logs, registration systems, third-party web analytics providers and ad-serving systems. Mostly these systems operate independently with a primary focus on describing what has happened. Through a deeper analysis, which will be enabled by the current effort, content providers will be able to use this data in more predictive ways. This in turn will allow content providers a more intelligent tool for serving higher-value content throughout the online experience. If successful, this will have implication for new rich media services and e-commerce.