Rashmi R. Sinha

Finding the flow in web site search

Designing a search system and interface may best be served (and executed) by scrutinizing usability studies.

The role of transparency in recommender systems

Recommender Systems act as a personalized decision guides, aiding users in decisions on matters related to personal taste. Most previous research on Recommender Systems has focused on the statistical accuracy of the algorithms driving the systems, with little emphasis on interface issues and the users perspective. The goal of this research was to examine the role of transprency (user understanding of why a particular recommendation was made) in Recommender Systems. To explore this issue, we conducted a user study of five music Recommender Systems. Preliminary results indicate that users like and feel more confident about recommendations that they perceive as transparent.

Empirically validated web page design metrics

A quantitative analysis of a large collection of expert-rated web sites reveals that page-level metrics can accurately predict if a site will be highly rated. The analysis also provides empirical evidence that important metrics, including page composition, page formatting, and overall page characteristics, differ among web site categories such as education, community, living, and finance. These results provide an empirical foundation for web site design guidelines and also suggest which metrics can be most important for evaluation via user studies.

Collection of dietary-supplement data and implications for analysis.

The role of nutrient intake in disease outcome is often examined in epidemiologic studies. Most such studies conducted in the United States, however, have not included fortified foods or vitamin supplements. In the United States, these are important sources of vitamins C and E, but not of beta-carotene. In addition, the importance of these nutrient sources varies by race, sex, and age. Failure to include these sources produces errors in nutrient estimates, notable misclassification of individuals with regard to their total intake, and rankings of intake that bear little or no relationship to blood concentrations of those nutrients. Implications for statistical analysis are also considered. Risk analyses in which nutrients from supplements are handled as control variables or are analyzed separately may impair the ability to detect associations between total nutrient intake and disease risk. The additional source of misclassification of nutrient status for vitamins C and E and other nutrients derived from fortified foods or supplements would make it more difficult to obtain significant and consistent results in etiologic studies of these nutrients; this has not been a factor for beta-carotene, which is derived almost exclusively from fruits and vegetables.

Persona development for information-rich domains

Designing information architecture for complex websites requires understanding user information needs and mental models in that domain. Personas, or user archetypes, created for such domains should also reflect types of information needs, and usage of information set. We have created a statistical technique to identify important underlying groupings of information needs. In a preliminary study, we show how designers can use this information in conjunction with data from interviews and observations to generate and refine personas.

Hierarchical faceted metadata in site search interfaces

One of the most pressing usability issues in the design of large web sites is that of the organization of search results. A previous study on a moderate-sized web site indicated that users understood and preferred dynamically organized faceted metadata over standard search. We are now examining how to scale this approach to very large collections, since it is difficult to present hierarchical faceted metadata in a manner appealing and understandable to general users. We have iteratively designed and tested interfaces that address these design challenges; the most recent version is receiving enthusiastic responses in ongoing usability studies.

Rapid information architecture prototyping

To create user-centered information architectures (IA), designers need a structured methodology that allows them to move rapidly from initial exploration of domain, to designing and testing information architecture. Additionally, in rapidly changing domains, design needs to flexibly incorporate future additions and evolutions. Finally, IA design should also take into account business concerns and goals. This paper describes Rapid Information Architecture Prototyping, a three-stage methodology for creating and testing IA based on user and business requirements. First, stakeholder analysis is used to understand business and organizational context, while free-listing exercises are used to explore the domain. Next, results of free-listing are used in an open card-sorting to understand user mental models and generate prototype IAs. Finally, closed card-sorting is used to evaluate and choose between candidate structures. The last two stages can be used in an iterative manner to design and test prototype IAs. The results yield a future-oriented IA that can flexibly incorporate future changes to site content and functionality, and provide design direction for years to come.