Web Personalisation Through Incremental Individual Profiling and Support-based User Segmentation

Abstract

Online personalised "my*" services are gaining popularity due to a growing customer need for information filtering and customisation. However, current systems mostly rely on some general usage and customer interaction in selecting components from prespecified blocks of content. The demand is great for high-quality unsupervised services on the customer side and for enabling techniques on the vendor side. Furthermore, individual profiles and, thus, personalised content should reflect changing individual behaviour. How do we efficiently build and maintain up-to-date personalised services for a large number of individuals? A compact and efficient, incrementally updatable representation of individual profiles is crucial. In addition, methods are required for efficient comparison of such profiles.Here we propose a methodology for building up-to-date personalised services. Individual profiles are represented as space-efficient prefix trees that are inherently easy to update incrementally. To measure the similarity of profiles, and also for the purpose of segmentation, we define a support-based metric that exploits the advantages of the tree-based structure. We evaluate our method on anonymised web data of 10,000 customers of an investment bank collected over 1.5 years.

Standing on the schemas of giants

“…Some work has looked at saving information from the search process so that others can re-use it in the future [1,37,41]. Research in personalized web search often uses the implicit traces of others' opinions of behaviors to train collaborative filtering systems to recommend items matching individuals' information seeking needs [15,16,19,27,49]. Similarly, social navigation uses the traces of others' behaviors to point people to useful places that others have found [38,13].…”

Inform Product Change through Experimentation with Data-Driven Behavioral Segmentation

“…Behavioral segmentation using distance-based clustering and probability-based mixture models has been extensively studied in the fields including user targeting in marketing [12], online advertising [13] [14], personalizing content serving [15], identifying user behavior on social networks [16], and modeling shared interests of e-commerce users [17]. Applying segmentation in experimentation was also discussed in previous literature.…”