Cross-device Web Search

Dan Wu, Jing Dong, Shaobo Liang

• Publisher: Routledge
• Year: 2022
• Print ISBN: 9780367193133
• eBook ISBN: 9780429513770
• Edition: 1
• Topic: Languages & Linguistics
• Subtopic: Library & Information Science
• Index: Languages & Linguistics

Cross-device Web Search is the first book to examine cross-device search behavior, which takes place when people utilize multiple devices and several sessions to research the same topic.

Providing a comprehensive examination of cross-device search behaviors, the book also models and analyses their most important features and, by doing so, helps to elucidate the motivations behind such behaviors. Drawing on a variety of methods and sources, including system design, user experiments, and qualitative and quantitative analysis, the book introduces cross-device search, relates it to relevant conceptual models, and identifies cross-device search topics. Providing discussion of a comprehensive range of behaviors in the context of cross-device search, including querying, gazing, clicking, and touching, the book also presents the design and development of a system to support cross-device search, explores cross-device search behavior modeling, and predicts users' search performance.

Cross-device Web Search will be of great interest to academics and students situated in the fields of library and information science, computer science, and management science. The book should also provide fascinating insights to practitioners and others interested in information search retrieval, information seeking behavior, and human-computer interaction communities.

1 Introduction to Cross-device Search

DOI: 10.4324/9780429201677-1

1.1 Background

With the development of smart devices, people have embraced not only mobile phones, desktops, laptops, and tablets, but also smart TVs, wearable watches, and other devices. Cross-device ownership refers to the fact that individuals own and use multiple digital devices from which they can access content, play games, watch TV, and carry out other activities (Connie, 2017). According to market data, 77% of online users own at least one mobile and one PC device, and 44% of online users own at least one smartphone, one tablet, and one PC device (Verto, 2017). Users’ human-computer interaction tasks will occur on different devices and has become a common phenomenon.

With increasing device functions and types, users’ network search methods are increasingly diversified. Under the environment of cross-device ownership, users’ search activities will span different types of devices and may switch among various devices. Users’ complex search tasks also may span a long time, often needing multiple search sessions, and occurring on different devices (Han, Zhen, & He, 2015). For instance, a person used his/her smartphone to search for information about a stock in the morning. Then, he/she came to the office and used the desktop to search for relevant financial news and to buy stocks. After dinner, he/she sits on the sofa at home and uses an iPad to search for stock price trends. This is a typical example of a cross-device search.

In the above example, the user’s search activity occurs on different devices, and a search task also occurs in multiple search sessions. Nowadays, many, the various options in search devices provide a diverse ecosystem of personal devices for cross-device search (Geronimo, Husmann, & Norrie, 2016). The analysis of logs from the United States local market showed that about 5% of users possess more than one device and that queries issued by multi-device users account for over 16% (Montañez, White, & Huang, 2014). Search across multiple devices by an individual has become a common usage pattern since people can query search providers almost any time and from anywhere (Wang, Huang, & White, 2013).

According to CNNIC data (2020), 99.7% of Chinese netizens use mobile phones to surf the Internet. The proportion of using desktop, laptops, TVs, and tablets to access the Internet is 32.8%, 28.2%, 24.0%, and 22.9%, respectively. Interacting with two or more devices is the norm for multi-device users, and using a mobile phone first does not mean the mobile phone is the only device used. The majority of multi-device users show a preference for using different devices for searching.

In previous studies (MacKay & Watters, 2008a; MacKay & Watters, 2008b; Morris, Ringel Morris, & Venolia, 2008; Sellen, Murphy, & Shaw, 2002), researchers investigated the cross-device search behavior of specific groups (such as knowledge workers, students, researchers, etc.), through a combination of interviews, log data, and other research methods. These studies show that cross-device searches will be triggered by various search topics and tasks, and searchers will use different methods to save information (such as bookmarks, notes, e-mail to themselves, etc.) before resuming the previous search process and continuing the search. However, these methods cannot help users search on different devices better. Karlson, Meyers, Jacobs, Johns, & Kane, (2009) also point out that tasks cannot be easily carried over between devices due to lack of support.

As stated, with the ubiquitous popularity of mobile devices, cross-device search is a current phenomenon all over the world; thus, the topics and the outcomes of this book are relevant not just to China but to the whole world. This book can fulfill various needs. To academics, this book will stimulate research interests regarding information retrieval, information-seeking behavior, and HCI communities. The findings about cross-device search behavior and the models we will develop can enrich existing research in the search domain. To internet industries, this book can provide implications on improving search surface and function design. To the general public who wants to know about cross-device search behavior, this book can provide a systematic and comprehensive introduction, including theoretical, technical, and empirical studies.

The topic of this book is innovative. Currently, there is no book concerning cross-device search behavior existing on the market. Therefore, this book is equipped with a unique competitive strength in that it should be the first about cross-device search behavior research.

1.2 Key Terms

In this part, we introduce several important key terms, which are closely related to the research topic. The key terms we list here will appear in this book. At the beginning of this book, we hope to introduce the key words to the readers and help them understand these concepts.

1.2.1 Cross-device/Cross-screen

As mentioned above, as users have more smart devices, they often interact on different devices over a period of time, and there is a certain correlation between these interactions on different devices. Of course, it can also be said that the users’ interactions are on different screens.

Neate, Jones, & Evans (2017) studied user interactions on smartphones and smart TVs, and defined multi-device systems as those which involve computing experiences that span two or more devices. Another related key term is cross-screen. In Google’s report (2012), they defined “cross-screen” as the usage of a second screen (or more than one screen) for related work sequentially or for related work simultaneously.

In this book, “cross-screen” means having or utilizing more than one screen at the same time. From the above, the concepts of cross-device and cross-screen are very similar.

1.2.2 Multi-session Search

Previous studies (Pavani & Teja, 2015) have defined the search session as “the series of both clicked and unclicked URLs from user click-through logs.” Jansen, Spink, & Kathuria (2006) classify the search session “from a contextual viewpoint as a series of interactions by the user toward addressing a single information need.” Church, Smyth, Cotter, & Bradley (2007) concluded that a search session is “a session where the user has engaged in at least some search activity.”

In users’ daily lives, some information needs will not be completed within a short time. According to the definition of a search session, users’ search activities may be distributed across multiple search sessions. MacKay & Watters (2008a) defined multi-session searches as those that have a specific and defined goal and that require more than one web session to complete. Kotov, Bennett, White, Dumais, & Teevan (2011) also found that there are complex tasks requiring many queries spanning multiple search sessions.

1.2.3 Cross-session Search

The search session has also been defined as “a sequence of queries issued by a single user within a specific time limit” (Boldi et al., 2008). Similar to the concept of multi-session search, a cross-session task consists of a series of queries that corresponds to a distinct high-level information need (Kotov et al., 2011).

There are often a series of search activities to achieve a single goal, but spread across multiple sessions. This can be regarded as either a multi-session search or cross-session search. In our book, multi-session and cross-session searches can be regarded as a series of queries corresponding to a distinct high-level information need that is segmented into short time units.

1.2.4 Cross-device Search

Facing a complex search task, searchers may use different devices to find information at different times and locations (Wu & Liang, 2018).

Wang, Huang, & White (2013) studied users’ cross-device search behavior using large-scale log data, and defined the cross-device search as a set of seven tuples, including search history, query before device transition, query after device transition, previous device, current device, search session before device transition and search session after device transition. Montañez et al. (2014) proposed a related concept, “device transition,”...

Table of contents

• Cover Page
• Half Title Page
• Title Page
• Copyright Page
• Contents
• List of Figures
• List of Tables
• Support Material
• 1 Introduction to Cross-device Search
• 2 Theories Relating to Cross-device Search
• 3 Cross-device Search Topics
• 4 Cross-device Search Support Techniques and Systems
• 5 Cross-device Search Behaviors
• 6 Cross-Device Search Behavior Modeling
• 7 Discussion and Conclusion
• Index