Contextualised Browsing in a Digital Library's Living Lab

Abstract

Contextualisation has proven to be effective in tailoring search results towards the users' information need. While this is true for a basic query search, the usage of contextual session information during exploratory search especially on the level of browsing has so far been underexposed in research. In this paper, we present two approaches that contextualise browsing on the level of structured metadata in a Digital Library (DL), (1) one variant bases on document similarity and (2) one variant utilises implicit session information, such as queries and different document metadata encountered during the session of a users. We evaluate our approaches in a living lab environment using a DL in the social sciences and compare our contextualisation approaches against a non-contextualised approach. For a period of more than three months we analysed 47,444 unique retrieval sessions that contain search activities on the level of browsing. Our results show that a contextualisation of browsing significantly outperforms our baseline in terms of the position of the first clicked item in the result set. The mean rank of the first clicked document (measured as mean first relevant -MFR) was 4.52 using a non-contextualised ranking compared to 3.04 when re-ranking the result lists based on similarity to the previously viewed document. Furthermore, we observed that both contextual approaches show a noticeably higher clickthrough rate. A contextualisation based on document similarity leads to almost twice as many document views compared to the non-contextualised ranking.

Analysis of search stratagem utilisation

“…Among the actions proposed by DLs, the five stratagems studied could be enhanced in a way to improve the user experience and to adapt the results to his/her interest. Personalised approaches could be provided based on the user's previous actions and the stratagems he/she performed (see an example of contextual browsing for DL in [13]). The improvement could be achieved, for instance, by presenting a better ranking of items for the user.…”

“…As future work, we need to go beyond log analysis and perform user studies in order to compare user feedback with the findings of this study. In addition, it will be beneficial to include personalised items based on the stratagems performed by the user in the search system's interface [13]. This allows comparing the usage and the impact of stratagem-based ranking compared to the standard ranking.…”

ConSTR: A Contextual Search Term Recommender

“…In this paper, we propose ConSTR, a novel Contextual Search Term Recommender (ConSTR) that integrates a twolayer source for recommendations: one layer suggests similar terms based solely on the similarity between the user's current search terms (query) and terms that are similar in an abstract space provided by an embedding model (word2vec, GloVe etc.). The second layer suggests terms on the basis of the user's entire interaction context [6]. The interaction context is a representation of the user's previous search activities in the current session, including all prior queries and keywords.…”

Data

“…They describe an alternative, based on inferring types of data and indexing columns as a unit, and demonstrate some improvements in early success especially when long captions are not available. New retrieval models are needed, models, moreover, that can be optimized with limited training and/or interaction data [3,5].…”