Latent User Models for Online River Information Tailoring

Abstract

This paper explores Natural Language Generation techniques for online river information tailoring. To solve the problem of unknown users, we propose 'latent models', which relate typical visitors to river web pages, river data types, and river related activities. A hierarchy is used to integrate domain knowledge and latent user knowledge, and serves as the search space for content selection, which triggers user-oriented selection rules when they visit a page. Initial feedback received from user groups indicates that the latent models deserve further research efforts.

Personalized Questions, Answers and Grammars: Aiding the Search for Relevant Web Information

“…There is an increasing interest in the combination of language and user model techniques to obtain personalized linguistic resources (Brusilovsky and Millán, 2007;Milosavljevic and Oberlander, 1998;Stock et al, 2007;Han et al, 2014).…”

Content Selection in Data-to-Text Systems: A Survey

“…• Lack of prior knowledge of the users: One of the most crucial issues in adaptation is the lack of prior knowledge of the users. This issue has been raised by several researchers, such as (Janarthanam, 2011;Han et al, 2014), to name a few. Previous approaches to tackling this issue include the use of latent User Models (Han et al, 2014), initial questionnaires to derive information by the user (Reiter et al, 1999) and tackling first-time users using multi-objective optimisation (Gkatzia et al, 2016b).…”

“…As a user's prior knowledge can change through interactions, they introduce dynamic user modelling which allows to update a User Model after interacting with the user. Han et al (2014) suggest the use of latent User Models to NLG. In this framework, instead of directly seeking the users' preferences or the users' knowledge through questionnaires, the UMs are inferred through "hidden" information derived from sources such as Google Analytics.…”

“…Adaptation goal (Stock et al, 2007) reason of visit, interest and change of background knowledge during the visit (Demberg and Moore, 2006) user preferences on flights: price, number of stops, airport location etc. (Janarthanam and Lemon, 2010) user's inferred prior knowledge (Han et al, 2014) latent variables users' preferences Big Five Personality traits (Mahamood and Reiter, 2011) stress levels (Dethlefs et al, 2014) user ratings (Gkatzia et al, 2016b) unknown users (from other users' ratings)…”