Aggregated search: a new information retrieval paradigm

Arlind Kopliku (1) , Karen Pinel-Sauvagnat (1) , Mohand Boughanem (1, 2)

1 IRIT-SIG - Systèmes d’Informations Généralisées
2 UT3 - Université Toulouse III - Paul Sabatier

Traditional search engines return ranked lists of search results. It is up to the user to scroll this list, scan within different documents and assemble information that fulfill his/her information need. Aggregated search represents a new class of approaches where the information is not only retrieved but also assembled. This is the current evolution in Web search, where diverse content (images, videos, ...) and relational content (similar entities, features) are included in search results. In this survey, we propose a simple analysis framework for aggregated search and an overview of existing work. We start with related work in related domains such as federated search, natural language generation and question answering. Then we focus on more recent trends namely cross vertical aggregated search and relational aggregated search which are already present in current Web search.

Mots clés: Information Extraction; Information Retrieval; Focused Retrieval; Result Aggregation; Vertical Search; Aggregated Search; Relational Search

Domaines: Théorie de l'information [cs.IT]; Recherche d'information [cs.IR]

Arlind Kopliku, Karen Pinel-Sauvagnat, Mohand Boughanem. Aggregated search: a new information retrieval paradigm. ACM Computing Surveys, 2014, vol. 46 (n° 3), pp. 1-31. ⟨10.1145/2523817⟩. ⟨hal-01123809⟩