How Much Tracking Is Necessary? - The Learning Curve in Bayesian User Journey Analysis

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Standard

How Much Tracking Is Necessary? - The Learning Curve in Bayesian User Journey Analysis. / Stange, Martin; Funk, Burkhardt.
Proceedings of the Twenty-Third European Conference on Information Systems. AIS eLibrary, 2015.

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Harvard

Stange, M & Funk, B 2015, How Much Tracking Is Necessary? - The Learning Curve in Bayesian User Journey Analysis. in Proceedings of the Twenty-Third European Conference on Information Systems. AIS eLibrary, 23rd European Conference on Information Systems - ECIS 2015, Münster, Germany, 26.05.15. https://doi.org/10.18151/7217484

APA

Stange, M., & Funk, B. (2015). How Much Tracking Is Necessary? - The Learning Curve in Bayesian User Journey Analysis. In Proceedings of the Twenty-Third European Conference on Information Systems AIS eLibrary. https://doi.org/10.18151/7217484

Vancouver

Stange M, Funk B. How Much Tracking Is Necessary? - The Learning Curve in Bayesian User Journey Analysis. In Proceedings of the Twenty-Third European Conference on Information Systems. AIS eLibrary. 2015 doi: 10.18151/7217484

Bibtex

@inbook{02f02f601bbf4d3c855ca7f8227751ad,
title = "How Much Tracking Is Necessary? - The Learning Curve in Bayesian User Journey Analysis",
abstract = "Extracting value from big data is one of today{\textquoteright}s business challenges. In online marketing, for instance, advertisers use high volume clickstream data to increase the efficiency of their campaigns. To prevent collecting, storing, and processing of irrelevant data, it is crucial to determine how much data to analyze to achieve acceptable model performance. We propose a general procedure that employs the learning curve sampling method to determine the optimal sample size with respect to cost/benefit considerations. Applied in two case studies, we model the users' click behavior based on clickstream data and offline channel data. We observe saturation effects of the predictive accuracy when the sample size is increased and, thus, demonstrate that advertisers only have to analyze a very small subset of the full dataset to obtain an acceptable predictive accuracy and to optimize profits from advertising activities. In both case studies we observe that a random intercept logistic model outperforms a non-hierarchical model in terms of predictive accuracy. Given the high infrastructure costs and the users' growing awareness for tracking activities, our results have managerial implications for companies in the online marketing field. ",
keywords = "Business informatics, Big Data, Online Marketing, User Journey Analysis, Learning Curve, Bayesian Models",
author = "Martin Stange and Burkhardt Funk",
year = "2015",
month = may,
day = "29",
doi = "10.18151/7217484",
language = "English",
isbn = "978-3-00-050284-2",
booktitle = "Proceedings of the Twenty-Third European Conference on Information Systems",
publisher = "AIS eLibrary",
address = "United States",
note = "23rd European Conference on Information Systems - ECIS 2015, ECIS conference 2015 ; Conference date: 26-05-2015 Through 29-05-2015",
url = "https://www.ercis.org/, http://www.ecis2015.eu/",

}

RIS

TY - CHAP

T1 - How Much Tracking Is Necessary? - The Learning Curve in Bayesian User Journey Analysis

AU - Stange, Martin

AU - Funk, Burkhardt

N1 - Conference code: 23

PY - 2015/5/29

Y1 - 2015/5/29

N2 - Extracting value from big data is one of today’s business challenges. In online marketing, for instance, advertisers use high volume clickstream data to increase the efficiency of their campaigns. To prevent collecting, storing, and processing of irrelevant data, it is crucial to determine how much data to analyze to achieve acceptable model performance. We propose a general procedure that employs the learning curve sampling method to determine the optimal sample size with respect to cost/benefit considerations. Applied in two case studies, we model the users' click behavior based on clickstream data and offline channel data. We observe saturation effects of the predictive accuracy when the sample size is increased and, thus, demonstrate that advertisers only have to analyze a very small subset of the full dataset to obtain an acceptable predictive accuracy and to optimize profits from advertising activities. In both case studies we observe that a random intercept logistic model outperforms a non-hierarchical model in terms of predictive accuracy. Given the high infrastructure costs and the users' growing awareness for tracking activities, our results have managerial implications for companies in the online marketing field.

AB - Extracting value from big data is one of today’s business challenges. In online marketing, for instance, advertisers use high volume clickstream data to increase the efficiency of their campaigns. To prevent collecting, storing, and processing of irrelevant data, it is crucial to determine how much data to analyze to achieve acceptable model performance. We propose a general procedure that employs the learning curve sampling method to determine the optimal sample size with respect to cost/benefit considerations. Applied in two case studies, we model the users' click behavior based on clickstream data and offline channel data. We observe saturation effects of the predictive accuracy when the sample size is increased and, thus, demonstrate that advertisers only have to analyze a very small subset of the full dataset to obtain an acceptable predictive accuracy and to optimize profits from advertising activities. In both case studies we observe that a random intercept logistic model outperforms a non-hierarchical model in terms of predictive accuracy. Given the high infrastructure costs and the users' growing awareness for tracking activities, our results have managerial implications for companies in the online marketing field.

KW - Business informatics

KW - Big Data

KW - Online Marketing

KW - User Journey Analysis

KW - Learning Curve

KW - Bayesian Models

U2 - 10.18151/7217484

DO - 10.18151/7217484

M3 - Article in conference proceedings

SN - 978-3-00-050284-2

BT - Proceedings of the Twenty-Third European Conference on Information Systems

PB - AIS eLibrary

T2 - 23rd European Conference on Information Systems - ECIS 2015

Y2 - 26 May 2015 through 29 May 2015

ER -

Links

DOI

Recently viewed

Publications

  1. Multi-view discriminative sequential learning
  2. Web-scale extension of RDF knowledge bases from templated websites
  3. Dispatching rule selection with Gaussian processes
  4. Homogenization methods for multi-phase elastic composites with non-elliptical reinforcements
  5. Towards a Bayesian Student Model for Detecting Decimal Misconceptions
  6. Foundations and applications of computer based material flow networks for einvironmental management
  7. Artificial Intelligence Algorithms for Collaborative Book Recommender Systems
  8. Learning from Erroneous Examples: When and How do Students Benefit from them?
  9. Study on the effects of tool design and process parameters on the robustness of deep drawing
  10. Adjustable automation and manoeuvre control in automated driving
  11. Backstepping-based Input-Output Linearization of a Peltier Element for Ice Clamping using an Unscented Kalman Filter
  12. Situated multiplying in primary school
  13. Oddih
  14. Performance of process-based models for simulation of grain N in crop rotations across Europe
  15. Passive Rotation of Rotational Joints and Its Computation Method
  16. Exploiting ConvNet diversity for flooding identification
  17. Denoising and harmonic detection using nonorthogonal wavelet packets in industrial applications
  18. Modellieren in der Sekundarstufe
  19. Making mutual learning tangible
  20. The effect of yield surface curvature change by cross hardening on forming limit diagrams of sheets
  21. Challenges for postdocs in Germany and beyond:
  22. Sustainable Consumption - Mapping the Terrain
  23. Implementing aspects of inquiry-based learning in secondary chemistry classes: a case study
  24. Integrating resilience thinking and optimisation for conservation
  25. An Integrative Framework of Environmental Management Accounting
  26. A robust model predictive control using a feedforward structure for a hybrid hydraulic piezo actuator in camless internal combustion engines
  27. Comparative study on the dehydrogenation properties of TiCl4-doped LiAlH4 using different doping techniques
  28. Evaluating a Bayesian Student Model of Decimal Misconceptions
  29. Design of Reliable Remobilisation Finger Implants with Geometry Elements of a Triple Periodic Minimal Surface Structure via Additive Manufacturing of Silicon Nitride
  30. Spectral Early-Warning Signals for Sudden Changes in Time-Dependent Flow Patterns
  31. Effect of gap distortion on the field splitting of collective modes in superfluid He3-B
  32. Formative assessment in inclusive mathematics education in secondary schools