Citation Context

...vior such as SNIF-ACT [28] to derive a theoretically grounded set of Web site design principles, or the use of information scent to predict surfing patterns of users looking for information on a site =-=[7]-=-. Although there are commonalities with information foraging research, our study specifically addresses behavioral variability not fully handled in the foraging literature. The very need for users to ...

Citation Context

...e is based on a concept called Information Scent, which is described next.s3. INFORMATION GOALS AND INFORMATION SCENT The ScentTrails technique is based on the theoretical notion of information scent =-=[8,26]-=- developed in the context of information foraging theory [27]. In this context, Information Foraging is related to other studies, such as Berrypicking [3] and ASK [4], of how users interleave directed...

Citation Context

...gage in. Beyond the Web, search is a central activity for users of corporate intranets, specialized databases (from shopping to Medline), and increasingly for personal archives of documents and email =-=[4]-=-. Given the importance and ubiquity of search, it is remarkable how similar most search interfaces are. Users typically type a few words into a query box and receive a rank-ordered list of search resu...

Citation Context

... on-line. Such “click-stream” studies collected on a large scale can reveal emergent patterns in Web use, showing for example navigational patterns or enabling mathematical modeling of Web use (e.g., =-=[7,10]-=-). Other studies using on-line data are of a task analytic nature, describing the moment-to-moment Volume No. 4, Issue No. 1 227sPaper: Web Behavior Patterns CHI changing the world, changing ourselves...

Citation Context

...chers have discovered that we can predict and simulate how users surf sites with specific tasks, such as how patients access their personal medical records and seek answers to their medical questions =-=[7, 4]-=-. One of these models is based on the notion of “Information Scent,” which is the user’s perception of the value and the cost of accessing a particular piece of information [6]. The idea is that the u...

Citation Context

...stimates of the transition probabilities between pairs of journals according to how frequently they were consecutively visited by users within the same session (see Bollen (2005,2008)[6, 5], related: =-=[10, 20, 19]-=-). The resulting usage network consisted of 3,617,368 connections, resulting in a network density of 6.3%. The usage network was represented as the 7, 575 × 7, 575 matrix U whose entries ui,j correspo...

Citation Context

...n a site by applying statistical analysis to user log data. Pitkow and Pirolli [27] predicted web pages that users are likely to request by applying data mining technique to user log data. Chi et al. =-=[6]-=- applied techniques used in information retrieval research to estimate the likelihood of selecting each link in a given website for given information needs. Except for Chi et al. [6], the common chara...

Citation Context

...problem by first extending our previous image attention model [3] to include a time constraint. We also mathematically formulate the optimal browsing path problem based on information foraging theory =-=[6]-=-[19] which has been used to analyze the trade-offs in the value of information gained against the costs of performing activity in human-computer interaction tasks. The novel contributions of this pape...

Citation Context

...rrypicking, but puts less emphasis on the changing information needs. Instead, it focuses more on proximal cues that direct the searcher to more interesting information items or navigational elements =-=[9, 39]-=-. However, both berrypicking and information scent describe a type of low-level navigation in which individual resources or navigational elements may guide the information seeker, but they do not prov...

Citation Context

...hown that only 10-20% of queries contained any advanced syntax [14][25]. This analysis can be a useful way of capturing characteristics of users interacting with IR systems. Research in user modeling =-=[6]-=- and personalization [30] has shown that gathering more information about users can improve the effectiveness of searches, but require more information about users than is typically available from int...