Search result re-ranking based on gap between search queries and social tags

Personalized Web Search Using Emotional Features

“…The emotion values of an individual page are calculated by looking up the emotion values of the words in the page from the emotion dictionary and averaging them 4 . In this way, a page has an emotion value ranging from 0 to 1, since the emotion values of the words in the emotion dictionary range from 0 to 1.…”

“…Bendersky et al [3] presented a passage-based approach to leverage information about the centrality of the document passages with respect to the initial search results list. Yan et al [4] proposed a Query-Tag-Gap algorithm to re-rank search results based on the gap between search queries and social tags. Tyler et al [5] utilized the prediction of re-finding (finding the pages that users have previously visited) to re-rank pages.…”

“…Many aspects are utilized to achieve effective re-ranking of initial search results, such as query logs [1], authorship [2], passage centrality [3], social tags [4], re-finding [5], multiple pairwise relationships between pages [6], temporal features [7], and demographical contexts (gender, age and income) [8]. Re-retrieval of new pages based on query expansion or reformulation is another effective strategy for improving retrieval accuracy, when initial search results in response to a query contain no pages relevant to users' search intentions.…”

Latent subject-centered modeling of collaborative tagging

“…A few prominent example domains include tag-based search [Bao et al 2007;Dmitriev et al 2006;Heymann et al 2008;Xu et al 2008;Yan et al 2009], tag-based clustering/classification [Heymann et al 2008;Lu et al 2009;Yin et al 2009;Zubiaga et al 2009], and tag-related recommender systems, including collaborative tag recommendation [Garg et al 2008;Jäschke et al 2008;Li et al 2010;Mishne 2006;Sigurbjörnsson et al 2008;Song et al 2011] and tag-aware item recommendation [Peng et al 2009[Peng et al , 2010Tso-Sutter et al 2008]. In addition, social bookmarking systems have also gained its popularity in many other areas, such as ontology learning [Mika 2007] and tag cloud visualization [Fujimura et al 2008].…”