Back to search

Cornell University

Information Design with Big Data

Abstract

dc:description.abstract

This dissertation consists of three essays that investigate the effects of information design from different aspects and in different business contexts. The first essay --- information design in storytelling --- studies the impact of information structure in the form of induced suspense and surprise on audience experience in the context of movie viewing, and identifies significant impacts of surprise on audience experience and evaluation. By formulating the information design problem as a constraint optimization problem as in Ely et al. (2015), we propose and validate textual measures of suspense and surprise based on a merged dataset of movie scripts, movie features, and market outcome statistics. The second essay --- cognitive categorization, memorability, and likability --- explores what makes a visual design memorable and likable by proposing and providing scalable methods to descriptively quantify and evaluate two cognitive processes for logos: (1) perceptual categorization; and (2) functional categorization. With a dataset consisting of 125,270 logo designs from the U.S. market, spanning 39 industry categories, and annotated scores of memorability and likability, we evaluate both the absolute and relative impacts of two forms of cognitive categorization and logo features on design memorability and design likability. In addition, we explore multiple methods for logo clustering and analyze drivers of memorability and likability. To validate the cognitive interpretation of our proposed measures, we further gather and incorporate human perceptual templates into the algorithm. We discuss managerial insights for logo design. Third essay reviews the existing research on the intersection with a focus on visual image data and proposes a classification scheme for the diverse range of computer vision methods and constructs for visual marketing that has been developed in the literature. The classification criteria include the nature of research questions, application contexts, computational methods, and forms of big data. Additionally, the paper provides comparative evaluations of each criterion both horizontally and vertically, a normative guide on the use of these systems and results under different situations, and an agenda for future research.

Degree

thesis:*
Name thesis:degree_name
Ph.D., Management
Level thesis:degree_level
Doctor of Philosophy
Discipline thesis:degree_discipline
Management
Grantor
Cornell University
Year dc:date.issued
2019

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Hu, Shengli
Chair dc:contributor.chair
  • Kadiyali, Vrinda
Committee members dc:contributor.committeemember
  • Mankad, Shawn Pankaj
  • Hariharan, Bharath

Subjects

dc:subject × 9

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Dc Identifier Bibid
11050649
Dc Identifier Other
ProQuest Submission ID: 11499
ProQuest Publication ID: 13898103

Chain of custody

source
Harvested from
Cornell University
Base URL
ecommons.cornell.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Hu, Shengli. Information Design with Big Data. Doctor of Philosophy thesis, Cornell University, 2019. https://hdl.handle.net/1813/67665