Search results

This study aims to investigate and compare mobile and desktop user search behaviours of the 1300K site, a Korean shopping search engine, by using transaction log analysis.

Transaction logs of 1300K site were collected over a three months’ period, from 1 January to 31 March 2018. The data set of this study consists of 1,149,690 desktop queries, 2,346,938 mobile queries, 2,481,747 desktop browsing activities and 2,550,309 mobile browsing activities. This study quantitatively analyses transaction log of 1300K site.

The results of this study show that mobile usage is higher than desktop usage: there are more mobile sessions than desktop sessions and the number of mobile queries is more than double of desktop queries. Overall, mobile query search behaviours are more simple, targeted and focused than desktop query search behaviours. Also, mobile browsing behaviours are more simple and passive than desktop browsing behaviours. However, mobile click behaviours are more active than desktop click behaviours.

To the best of the authors’ knowledge, this study appears to be the first of its type in Korea that compared search behaviours of a large number of users on desktop computers and mobile phones. To identify various characteristics of user search behaviours, this study analyses users’ directory browsing behaviour and click behaviour as well as query search behaviour. The results of this study can be implemented to address the effective improvement and development of search services and interfaces for different devices.

This paper discusses the differences in search pattern transitions for mobile phone, tablet and desktop devices by mining the transaction log data of a library online public access catalogue (OPAC). We aimed to analyze the impacts of different devices on user search behavior and provide constructive suggestions for the development of library OPACs on different devices.

Based on transaction logs which are 9 GB in size and contain 16,140,509 records of a university library OPAC, statistics and clustering were used to analyze the differences in search pattern transitions on different devices in terms of two aspects: search field transition patterns and query reformulation patterns.

Search field transition patterns are influenced by the input function and user interfaces of different devices. As reformulation times increase, the differences between query reformulation patterns among different devices decrease.

Mobile-side libraries need to optimize user interfaces, for example by setting web page labels and improving input capabilities. Desk-side libraries can add more suggestive content on the interface.

Unlike previous studies, which have analyzed web search, this paper focuses on library OPAC search. The search function of mobile phones, tablets and desktops were found to be asymptotic, which was a good illustration of how devices have a large impact on user search behavior.

Mobile users increasingly employ location-based map searches in their daily lives. However, it is still relatively unknown about mobile users’ map related search behaviors. The purpose of this paper is to discover the interactions between the users and mobile map search systems, to reveal the shortcomings of existing mobile map search functions, and to propose improvement suggestions.

Based on a set of controlled user experiments performed on the Baidu mobile phone map, this paper empirically examines users’ location-based mobile search behaviors, such as timing, metering, judging and so on. This paper also conducts statistical correlation tests to generate relation tables and diagrams regarding each variable, for example, the relation between the retrieval time and the retrieval steps.

The results indicate that mobile map users have two important characteristics in their search behaviors: first, mobile map users always follow the single search path. Second, the mobile map search efficiency of users is always low.

The situation simulation testing method is mainly used for the construction of a mobile information search behavior environment, which may make the users be nervous and have some effect on the search efficiency.

Based on the identification of user behaviors, this paper provides suggestions to optimize and improve mobile map search systems.

This paper studies users’ mobile map search behavior based on location and explores the features of user behavior from the perspective of human-computer interaction.

Web search is more and more moving into mobile contexts. However, screen size of mobile devices is limited and search engine result pages face a trade-off between offering informative snippets and optimal use of space. One factor clearly influencing this trade-off is snippet length. The purpose of this paper is to find out what snippet size to use in mobile web search.

For this purpose, an eye-tracking experiment was conducted showing participants search interfaces with snippets of one, three or five lines on a mobile device to analyze 17 dependent variables. In total, 31 participants took part in the study. Each of the participants solved informational and navigational tasks.

Results indicate a strong influence of page fold on scrolling behavior and attention distribution across search results. Regardless of query type, short snippets seem to provide too little information about the result, so that search performance and subjective measures are negatively affected. Long snippets of five lines lead to better performance than medium snippets for navigational queries, but to worse performance for informational queries.

Although space in mobile search is limited, this study shows that longer snippets improve usability and user experience. It further emphasizes that page fold plays a stronger role in mobile than in desktop search for attention distribution.

The paper shows how access to knowledge can be enhanced by using a set of innovative approaches and technologies based on the semantic web.

Emerging trends in knowledge access are considered followed by a description of how ontologies and semantics can contribute. A set of tools is then presented which is based on semantic web technology. For each of these tools a detailed description of the approach is given together with an analysis of related and future work as appropriate.

The tools presented are at the prototype stage but can already show how knowledge access can be improved by allowing users to more precisely express what they are looking for and by presenting to them in a form that is appropriate to their current context.

The tools show promising results in improving access to knowledge which will be further evaluated within a practical setting. The tools will be integrated and trialled as part of case studies within the SEKT project. This will allow their usability and practical applicability to be measured.

Ontologies as a form of knowledge representation are increasing in importance. Knowledge management, and in particular knowledge access, will benefit from their widespread acceptance. The use of open standards and compatible tools in this area will be important to support interoperability and widespread access to disparate knowledge repositories.

The paper presents research in an emerging but increasingly important field, i.e. semantic web‐based knowledge technology. It describes how this technology can satisfy the demand for improved knowledge access, including providing knowledge delivery to users at the right time and in the correct form.

State-of-the-art cloud applications are problematic for collaborative document management; their current design does not encourage active personal folder categorization. Cloud applications such as Google Drive and Microsoft’s OneDrive store documents automatically, so at no point are users directed to categorize them by placing them in folders. To encourage active categorization and promote effective retrieval of cloud documents, the authors designed an add-on “nudge” called Personal Organizer which prompts Google Drive users to categorize by storing cloud documents in personal folders. The add-on prompt is triggered when users attempt to close uncategorized or unnamed documents. The purpose of this paper is to test whether using the Personal Organizer add-on leads participants to actively store their documents in folders that they personally created, and whether this promotes more successful and efficient retrieval.

To test the add-on, the authors conducted a pretest-manipulation-post-test intervention study with 34 participants lasting over three months. In both tests, participants were asked to retrieve personal documents taken from their own “Recents” list to improve ecological validity.

Using our add-on doubled the percentage of documents that were actively stored in folders. Additionally, using personally created folders substantially improved retrieval success while decreasing retrieval time.

Implementing our findings can improve document storage and retrieval for millions of users of collaborative cloud storage. The authors discuss broader theoretical implications concerning the role of active organization for retrieval in collaborative repositories, as well as design implications.

Previous research has repeatedly shown that people only search for files in a small minority of cases when they do not remember the file's location. The current study aimed to examine whether there is a group of hyper-searchers who search significantly more than others. Based on previous neurocognitive studies, this study aims to hypothesize that if such a group exists, they will have superior verbal memory and reduced visuospatial memory.

In total, 65 participants completed a questionnaire estimating their search percentages, as well as reporting demographic data. Verbal memory was measured using the Wechsler logical memory test, and visuospatial memory was assessed using an online card memory game.

Hyper-searchers were defined as participants with search percentage of over one standard deviation (SD) above the mean. The average search percentage of the seven participants who met this criterion was 51% (SD = 14%), over five times more than the other participants (M = 10%, SD = 9%). Similar results were obtained by re-analyzing data from four previous papers (N = 1,252). The results further confirmed the hypothesis that hyper-searchers have significantly better verbal memory than other participants, possibly making searching easier and more successful for them. Lastly, the search percentage was positively predicted by verbal memory scores and negatively predicted by visuospatial memory scores. Explanations and future research are discussed.

This preliminary study is the first to introduce the concept of hyper-searchers, demonstrate its existence and study its causes.