11/16/2023
By Xavier Babu

The Department of Operations and Information Systems at the Manning School of Business invites you to attend a doctoral dissertation proposal by Xavier Babu on “Systematic Affinity Analytics and Analytics Application Platform Design Research for Digital Business Analytics.”

Date: Wednesday, Nov. 29, 2023
Time: 12:30 to 2:30 p.m.
Location: Via Zoom 
Thesis/Dissertation Title: Systematic Affinity Analytics and Analytics Application Platform Design Research for Digital Business Analytics

Committee Members:

  • Julie Zhang (Co-chair), Ph.D., Department of Operations & Information Systems, Manning School of Business, UMass Lowell
  • Luvai Motiwalla (Co-chair), Ph.D., Department of Operations & Information Systems, Manning School of Business, UMass Lowell
  • Berk M Talay, Ph.D., Department of Marketing Entrepreneurship and Innovation, UMass Lowell
  • Xiaobai (Bob) Li, Ph.D., Department of Operations & Information Systems, Manning School of Business, UMass Lowell

Abstract:
This dissertation proposal centers on systematic affinity analytics (theoretical) research examining the affinities and the dynamics among product reviews, managers' responses, reputation and convergence or divergence using contextual-based affinities by employing a deep learning framework. Secondly, it focuses on systematic analytics application platform design (practical) research using Action Design Research (ADR), a Design Science Research (DSR) genre, to help analytical professionals.

Systematic Affinity Analytics (theoretical) Research: We study the contextual-based affinities between managers' responses to product reviews, managers' responses with their peers, and focal product reviews with reviews from rivals, along with the impact of affinities on product reputation and review convergence. While previous research has explored the influence of reviews on product reputation, little attention has been given to the effectiveness of manager response and the contextual-based affinities in this process. This study fills this gap by examining the affinities and the dynamics among product reviews, managers' responses, reputation and convergence or divergence. With a travel dataset from TripAdvisor.com, we develop a systematic affinity analytics research model by employing a deep learning framework to understand semantic textual information in managers' responses and reviews and utilizing Panel Vector Autoregression models to examine the dynamic relationships between managers' responses and product reviews, the interactions between managers and their peers, and the interplay of focal product reviews with rivals. Through our study and new systematic affinity analytics research contribution, we provide a catalyst for more research in product business management using the contextual-based research model we developed to analyze new hypotheses, primarily creating a comparable system with the peers' data.

Systematic Analytics Application Platform Design (practical) Research: Digital marketing analytics teams are generally expected to have primarily analytical programming skills. It is unusual for the teams to be asked with non-analytical programming, UI, and other IT skills. Yet, they must develop efficient analytical models and applications with IT platforms that require developing applications with high-quality UI, orchestration, security, and privacy. This requirement motivated us to design a systematic Digital Marketing Analytics Application Platform (DMAAP) using Action Design Research (ADR), a Design Science Research (DSR) genre, to help analytical professionals. First, it contributes to the descriptive knowledge concerning the problem space, namely, the development of DMAAP key aspects, by identifying the platform's requirements through rigorous research and input from the implementation team. Second, it contributes to the design science research, which we named "extended ADR", by proposing a new personalization stage adapted from Mullarkey & Hevner (2019). Third, it contributes to prescriptive knowledge concerning the solution space, namely, DMAAP design principles and the instantiated artifact for DMAAP design and framework. The final contribution is the field study to evaluate the DMAAP instantiation by applying relevant digital-marketing business use cases that showcase the artifact, the DMAAP, and in-depth implementation in a digital marketing firm. The field study, which applied ADR-DSR methodology, was instrumental in continually improving the DMAAP design. It was conducted with the digital marketing professionals at Epsilon, Inc. In sum, this project provides a forum for discussing the challenges of implementing the DMAAP platform and gaining insights into using DSR to overcome the challenges. Through our study, artifacts contribution and evaluation via on-the-ground research, we wish to serve as a springboard for extended research that adds to the discourse to design a broad class of personalized digital analytics application platforms, advancing sustainable digital data analytics for digital corporations.

The first essay (Chapter 1) focuses on developing a systematic affinity analytics research model by employing a deep learning framework to understand semantic textual information in managers' responses and reviews and utilizing Panel Vector Autoregression models to examine the dynamic relationships between managers' responses and product reviews, the interactions between managers and their peers, and the interplay of focal product reviews with rivals.

The second essay (Chapter 2) focuses on systematic analytics application platform design using Active Design Research to answer the research question: What are the appropriate design principles for a digital marketing analytics application platform (DMAAP) to improve the efficiency and effectiveness of analytics professionals, agnostic of the programming and analytical skills on digital-marketing platforms? Besides, it concentrates on rigorous research and field study evaluation by co-creating the instantiation of the DMAAP artifacts at Epsilon, headquarters in the USA. It is the data-tech platform and a division of Publicis Groupe, a French multinational advertising and public relations company.

All interested students and faculty members are invited to attend.