Technical Guide to Personalisation
Personalisation in a digital environment is complex. Evolving statistical methodologies, algorithmic machine learning and the advancements in Big Data compound the complexity.
There are however, best practices to getting the right results at the right time that deliver personalisation – consistently and accurately every time.
This paper explores the core methodologies behind personalisation and reviews the techniques that deliver those exacting results.
In this paper you’ll discover:
- How Naïve Bayes classifications provide insight into preference
- The principles behind how tags categorise and refine information
- Personal taste indicators based on purchase history, response or interest
- How Matrix Factorisation and latent semantic indexing are used for predications
- The best practice approaches to modelling data for personalisation and recommendations