Information Foraging Theory as a Form of Collective Intelligence for Social Search
This item is available under a Creative Commons License for non-commercial use only
The world wide web is growing in size and with the proliferation of large-scale collaborative computing environments Social search has become increasingly important. The focal point of this recent field is to assign relevance and trustworthiness to web-pages by taking into account the reader's perspective rather than web-masters point of view. Current web-searching technologies tend to rely on explicit human recommendations in part because it is hard to obtain user feedback. however these methods are hard to scale. Implicit feedback techniques are a potentially useful alternative. The challenge is in producing implicit web-rankings by reasoning over users' activity during a web-search but without recourse to explicit human intervention. This paper focuses on a novel Social Search formal model based on Information Foraging Theory, showing a different way to implicitly judge web entities by considering effort expended by users in viewing them. 100 university students were recruited to explicitly evaluate the usefulness of 12 thematic web-sites and an experiment was performed implicitly gathering their web-browsing activity. Correlation indexes were adopted and encouraging results where obtained suggesting the existence of a considerable relationship between explicit feedback and implicit derived judgements. Furthermore, a comparison of the results obtained and the results provided by Google was performed. The proposed nature-inspired approach shows that, by considering the same searching query, Social search to be more effective than the Google Page-Rank Algorithm. This evidence supports the presentation of a novel general schema for a Social search engine generating implicit web-rankings by taking into account the Collective Intelligence emerged from users by reasoning on their behaviour.
Luca, L., Barrett, S., Pierpaolo, D. (2009). Information foraging theory as a form of collective intelligence for social search. Computational Collective Intelligence, Semantic Web, Social Networks and Multiagent Systems, p.63-74