TY - JOUR
T1 - Don't Need All Eggs in One Basket
T2 - Reconstructing Composite Embeddings of Customers from Individual-Domain Embeddings
AU - Unger, Moshe
AU - Li, Pan
AU - Sen, Sahana Shahana
AU - Tuzhilin, Alexander
N1 - Publisher Copyright:
© 2023 Copyright held by the owner/author(s). Publication rights licensed to ACM.
PY - 2023/3/13
Y1 - 2023/3/13
N2 - Although building a 360-degree comprehensive view of a customer has been a long-standing goal in marketing, this challenge has not been successfully addressed in many marketing applications because fractured customer data stored across different "silos"are hard to integrate under "one roof"for several reasons. Instead of integrating customer data, in this article we propose to integrate several domain-specific partial customer views into one consolidated or composite customer profile using a Deep Learning-based method that is theoretically grounded in Kolmogorov's Mapping Neural Network Existence Theorem. Furthermore, our method needs to securely access domain-specific or siloed customer data only once for building the initial customer embeddings. We conduct extensive studies on two industrial applications to demonstrate that our method effectively reconstructs stable composite customer embeddings that constitute strong approximations of the ground-truth composite embeddings obtained from integrating the siloed raw customer data. Moreover, we show that these data-security preserving reconstructed composite embeddings not only perform as well as the original ground-truth embeddings but significantly outperform partial embeddings and state-of-the-art baselines in recommendation and consumer preference prediction tasks.
AB - Although building a 360-degree comprehensive view of a customer has been a long-standing goal in marketing, this challenge has not been successfully addressed in many marketing applications because fractured customer data stored across different "silos"are hard to integrate under "one roof"for several reasons. Instead of integrating customer data, in this article we propose to integrate several domain-specific partial customer views into one consolidated or composite customer profile using a Deep Learning-based method that is theoretically grounded in Kolmogorov's Mapping Neural Network Existence Theorem. Furthermore, our method needs to securely access domain-specific or siloed customer data only once for building the initial customer embeddings. We conduct extensive studies on two industrial applications to demonstrate that our method effectively reconstructs stable composite customer embeddings that constitute strong approximations of the ground-truth composite embeddings obtained from integrating the siloed raw customer data. Moreover, we show that these data-security preserving reconstructed composite embeddings not only perform as well as the original ground-truth embeddings but significantly outperform partial embeddings and state-of-the-art baselines in recommendation and consumer preference prediction tasks.
KW - 360-degree view of customer
KW - Deep Learning
KW - composite customer embedding
KW - customer preference prediction
UR - http://www.scopus.com/inward/record.url?scp=85152624482&partnerID=8YFLogxK
U2 - 10.1145/3578710
DO - 10.1145/3578710
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:85152624482
SN - 2158-656X
VL - 14
JO - ACM Transactions on Management Information Systems
JF - ACM Transactions on Management Information Systems
IS - 2
M1 - 3578710
ER -