Article Preview
Top1. Introduction
We are living in the age of big data, advanced analytics, and human information interaction (HII). The trend of ‘big data’ or ‘data deluge’ is evident in every walk of life. The tremendous growth of ‘the data’ not only triggered remarkable hype and buzz but presents enormous challenges that, in turn, offers incredible innovation and opportunity. An individual is exposed to a lot of data, which carries potential, insights, and knowledge. A information-seeker observes and learns, primarily for survival, and for societal and personal growth. Though, extracting relevant information is a multifaceted task, particularly to a naive user with uncertain information needs.
Consequently a user opportunistically, adapts several strategies, ranging from analytical search, bibliographical search, search by analogy, browsing, etc. to gain knowledge. Traditionally, two alternatives exist for information search, first prefer complex query language, e.g. SQL and second require interactive queries to synthesize information over information system. The former generally requires the uncommon expertise of data retrieval language, while the latter requires extensive manual efforts (Gleicher, Albers, Walker, Jusufi, Hansen & Roberts, 2011; Janiszewski, 1998). In both, the inherent user-system conversations are of exploratory nature and primarily based on prior understanding of information needs. Therefore, the information-seeking becomes discovery-oriented and illustration of extracted information play pivotal role.
Information visualization can turn a difficult relevance based information assessment task into effortless recognition task. As visual portrayal of data (i.e. content, semantics, correlations, etc.) is easy to perceive and reduces cognitive-effort. Most of the modern search systems/strategies are augmented with visualizations. In this, browsing and analytical searches are the two most cited for exploratory-seeking, and benefited from visual support (Kucher & Kerren, 2015). Designing visual interface to assist the exploratory search is growing practice in recent years (Reda, Johnson, Papka & Leigh, 2016). The multi-dimensional interactive visualization offers capabilities to peek into different aspects of retrieved data. Visual analytics tools offer displays, to understand the collection as a whole, discover meaningful hidden relationships, and formulate insights with a reduce effort (Pike, Stasko, & O'connell, 2009; Gan, Zhu, Liang, Cao & Zhou, 2014).
Simple information visualization has proven a useful tool to enable the user to explore the information, i.e. data analysis, query suggestion, recommendation/ prediction, etc. (Idreos, Papaemmanouil, & Chaudhuri, 2015; Graham & Kennedy, 2010; Skeels, Lee, Smith & Robertson, 2010; Borovina & Ferreira, 2017). The cognitive-science theory established that user finds recognitions easier than recall, as it is easier for a human to perceive something they see than described for the search. The approach motivates a rapid evolution for the design of visualization based tools, with aim to assist the users (Eppler, 2006; Alencar, de Oliveira & Paulovich, 2012; Levie, 1987; Sedlmair, Isenberg, Baur & Butz, 2011; Bleifuß, Bornemann, Johnson, Kalashnikov & Srivastava, 2018; Goldsmith, 1984).
In general, a human can process specific visual properties and the choice of visual representation. Current visualization technology supports a variety of search-interactions, to aid the user to progressively explore in ‘data deluge’. The exploratory conversations provide an active assistant to extraction and possible actions, to help the user and activities may entail a wide range of visualization talks. Though, it is hard for a seeker to look for specific tasks or visual metaphors. Primarily, due to the lack of comprehensive evaluation of visual features of visualization tools. Thus, it is unclear what role, if any, information visualization plays in the user search.