Location and time do matter: A long tail study of website requests

Kumar, Chetan; Norris, John B.; Sun, Yi

doi:10.1016/j.dss.2009.04.015

Cited by 15 publications

(3 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We now describe typical web graph characteristics by discussing the sizes of L, R, c and a in practice. As noted before, in most websites, a small number of 'head' pages contribute to a significant amount of the traffic while a long tail of the remaining pages contribute to the rest [8,13,17].…”

Section: The Recommendation Subgraph Problemmentioning

confidence: 73%

Recommendation Subgraphs for Web Discovery

Antikacioglu

Ravi

Sridhar³

2015

Proceedings of the 24th International Conference on World Wide Web

View full text Add to dashboard Cite

Recommendations are central to the utility of many websites including YouTube, Quora as well as popular e-commerce stores. Such sites typically contain a set of recommendations on every product page that enables visitors to easily navigate the website. Choosing an appropriate set of recommendations at each page is one of the key features of backend engines that have been deployed at several e-commerce sites. Specifically at BloomReach, an engine consisting of several independent components analyzes and optimizes its clients websites. This paper focuses on the structure optimizer component which improves the website navigation experience that enables the discovery of previously undiscovered content. We begin the paper by formalizing the concept of recommendations used for discovery. We formulate this as a natural graph optimization problem which in its simplest case, reduces to a bipartite matching problem. In practice, solving these matching problems requires superlinear time and is not scalable. Also, implementing simple algorithms is critical in practice because they are significantly easier to maintain in a production software package. This motivated us to analyze three methods for solving the problem in increasing order of sophistication: a local random sampling algorithm, a greedy algorithm and a more involved partitioning based algorithm. We first theoretically analyze the performance of these three methods on random graph models characterizing when each method will yield a solution of sufficient quality and the parameter ranges when more sophistication is needed. We complement this by providing an empirical analysis of these algorithms on simulated and real-world production data. Our results confirm that it is not always necessary to implement complicated algorithms in the real-world. Indeed, our results demonstrate that very good practical results can be obtained by using simple heuristics that are backed by the confidence of concrete theoretical guarantees.

show abstract

Section: The Recommendation Subgraph Problemmentioning

confidence: 73%

Recommendation Subgraphs for Web Discovery

Antikacioglu

Ravi

Sridhar³

2015

Proceedings of the 24th International Conference on World Wide Web

View full text Add to dashboard Cite

show abstract

“…The drawback is that these conclusions about their websites evaluated over time depended on observation and estimation of the GA dashboard rather than accurate computation. Because of this, interpretations of visitor's traffic data may be oversimplified [17] and subject to limitations endemic to Google Analytics' existing functionalities…”

Section: Introductionmentioning

confidence: 99%

Decision Support Based on Time-Series Analytics: A Cluster Methodology

Xing

Guo

Lowrance

et al. 2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Web analytic techniques have become increasingly popular, particularly Google Analytics time-series dashboards. But interpretations of a website's visits traffic data may be oversimplified and limited by Google Analytics existing functionalities. This means website mangers have to make estimations rather than mathematically informed decisions. In order to gain a more precise view of longitudinal website visits traffic data, the researchers mathematically transformed the existing Goggle Analytics' log data allowing the vectors of website visits per each year to be considered simultaneously. The methodology groups the data of an example website gathered over an 'x' year period into 'y' clusters of data. The results show that the transformed data is richer, more accurate and informative, potentially allowing website managers to make more informed decisions concerning promoting, developing, and maintaining their websites rather than relying on estimations.

show abstract

“…Moreover, the variables student motivation and student satisfaction comprise comprehensive, multi-item constructs in order to better understand student behavior. That is, customers may be segmented according to certain demographic characteristics, and administrators may use this information to understand customer satisfaction and predict certain customer behaviors, and accordingly create better strategies (Andrews, Brusco, & Currim, 2010;Kumar, Norris, & Sun, 2009). Ultimately, the final model was tested using structural equations modeling (SEM).…”

Section: Introductionmentioning

confidence: 99%

Understanding Critical Distance Learning Issues

Callaway

Alflayyeh

2011

Information Resources Management Journal

View full text Add to dashboard Cite

Distance education has been the topic of a substantial amount of research. However, prior studies have shown mixed results when trying to determine if a difference exists in student satisfaction between students in distance courses versus traditional courses. Prior empirical studies have been too narrow in scope, and a more comprehensive model is needed to better explain the factors influencing student satisfaction. Therefore, the current study includes student demographic factors, comprehensive measures of student motivation, and course format, as well as specific course features included, to fully explain student satisfaction. Structural equation modeling is used to test the model. Results indicate a positive association between demographics and motivation, between motivation and course format, between one demographic factor and course format, between course format and preferred features, between course format and satisfaction, and between course features and satisfaction.

show abstract

Location and time do matter: A long tail study of website requests

Cited by 15 publications

References 16 publications

Recommendation Subgraphs for Web Discovery

Recommendation Subgraphs for Web Discovery

Decision Support Based on Time-Series Analytics: A Cluster Methodology

Understanding Critical Distance Learning Issues

Contact Info

Product

Resources

About