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

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

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.

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

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, Deutschland, 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

Zuletzt angesehen

Aktivitäten

  1. Project Workshop on "Worker Flows, Match Quality, and Productivity" - 2019
  2. 13th IFAC and IEEE Conference on Programmable Devices and Embedded Systems - PDeS 2015
  3. Applied Econometrics with Stata for PhD Students
  4. Princeton University
  5. Thinking of Time - A Resource which Should be Allocated Equally
  6. Changing learning environments at university? Comparing the learning strategies of non-traditional European students engaged in lifelong learning.
  7. Predicting negotiation success with a multitude of negotiators’ inter-individual differences—a latent personality model of the successful negotiator
  8. Visualizing and analyzing big data sets: Results from the Student Bodies-Eating Disorders study
  9. "Crowds and Party" Reading Workshop with Jody Dean - 2019
  10. European University Institute
  11. Where is language use in the description of the Englishes? - ESSE 2006
  12. Comfort and Adaptive Cruise Control in Highly Automated Vehicles
  13. Panel Cointegration Testing with Time Trend and Analysis of Money Demand in OECD Countries
  14. Navigating Educational Ambidexterity: Exploring Leader-Member Exchange and Open Innovation
  15. Digital Capitalism meets Leberkaspeppi: Temporal Orientations in Business Models as a Source of Platform Power in Mature Industries
  16. THE EUROPEAN CONFERENCE ON MACHINE LEARNING & PRINCIPLES AND PRACTICE OF KNOWLEDGE DISCOVERY IN DATABASES 2017
  17. Ghosting the City – Zooming in on Otherwise Publics in Virtual Worlds
  18. 6th Workshop on Unintended Consequences
  19. Internet-based guided self-help to reduce depressive symptoms in teachers: Results from a randomized controlled trial
  20. Modelling biodegradability based on OECD 301D data for the design of mineralising ionic liquids
  21. Mathematical and Computational Applications (Fachzeitschrift)
  22. Response to Philip Hogh: Suffering, Pain and the Idea of Progress

Publikationen

  1. Cue predictability changes scaling in eye-movement fluctuations
  2. Service Level Driven Stock Allocation
  3. Digging into the roots
  4. Design for Product Care—Development of Design Strategies and a Toolkit for Sustainable Consumer Behaviour
  5. The effect of structural complexity on large mammal occurrence in revegetation
  6. Towards Advanced Learning in Dispatching Rule-Based Scheuling
  7. Integrating inductive and deductive analysis to identify and characterize archetypical social-ecological systems and their changes
  8. "Introduction," communication +1
  9. What Makes for a Good Theory? How to Evaluate a Theory Using the Strength Model of Self-Control as an Example
  10. Embedding Evidence on Conservation Interventions Within a Context of Multilevel Governance
  11. Drafts in Action
  12. Development and comparison of processing maps of Mg-3Sn-1Ca alloy from data obtained in tension versus compression
  13. Modeling of temperature- and strain-driven intermetallic compound evolution in an Al-Mg system via a multiphase-field approach with application to refill friction stir spot welding
  14. Discriminative clustering for market segmentation
  15. From Open Access to Open Science
  16. How to move the transition to sustainable food consumption towards a societal tipping point
  17. Integrating business models and enterprise architecture
  18. Concepts, Formats, and Methods of Participation
  19. Effective digital practice in the competence-oriented English as a foreign language classroom in Germany
  20. Trajectory tracking using MPC and a velocity observer for flat actuator systems in automotive applications