Characteristics of Dataset Retrieval Sessions: Experiences from a Real-life Digital Library

Abstract

Secondary analysis or the reuse of existing survey data is a common practice among social scientists. Searching for relevant datasets in Digital Libraries is a somehow unfamiliar behaviour for this community. Dataset retrieval, especially in the social sciences, incorporates additional material such as codebooks, questionnaires, raw data files and more. Our assumption is that due to the diverse nature of datasets, document retrieval models often do not work as efficiently for retrieving datasets. One way of enhancing these types of searches is to incorporate the users' interaction context in order to personalise dataset retrieval sessions. As a first step towards this long term goal, we study characteristics of dataset retrieval sessions from a real-life Digital Library for the social sciences that incorporates both: research data and publications. Previous studies reported a way of discerning queries between document search and dataset search by query length. In this paper, we argue the claim and report our findings of an indistinguishability of queries, whether aiming for a dataset or a document. Amongst others, we report our findings of dataset retrieval sessions with respect to query characteristics, interaction sequences and topical drift within 65,000 unique sessions.

Large-scale analysis of query logs to profile users for dataset search

“…They concluded their study by drawing clear differences between the search for dataset compared to publication; firstly, the average number of record views in dataset search sessions was significantly lower compared to sessions searching for publications. Secondly, similar to Kacprzak et al (2018), they reported difficulty in understanding the users' intent and their differences between dataset search and publication search, due to the minor differences between the two sets of queries (Carevic et al, 2020).…”

“…Nevertheless, in both of the setting having individuals searching for datasets, neither set necessarily represented how individuals would like to search for data. Carevic et al (2020) focused on characterising the users' intents when performing publication (document) search and research datasets in a Digital Library for the social sciences discipline. They studied all the user log data for an entire year, and extracted sessions that searched only for datasets and publications.…”

“…Majority of existing research in the literature focuses on user behaviour for searching textual documents/ web pages, images or videos. Limited research but growing interest exists in the research community to uncover user dataset search behaviour, in light of the vast amount of datasets that is becoming available on the Web due to the Open Data initiatives (Carevic et al, 2020). Similarly, we have seen growing interests in designing information retrieval models specifically for dataset search (Chapman et al, 2020;Brickley et al, 2019).…”

“…Dataset search is identified to be more challenging in comparison to classical information search. This is mainly due to the diverse and distinctive nature of dataset search that embeds both users' complex information need as well as query formulation (Carevic et al, 2020). Additionally, dataset search involves more difficult selection decision compared to search for information (Kern and Mathiak, 2015).…”

“…To date, a few studies exist that have attempted to understand the dataset users' behaviour and intentions (Kacprzak et al, 2019;Carevic et al, 2020;Chen et al, 2019). These studies however remain limited in explaining various aspects of the users' behaviour.…”