Machine Learning and Consumer Data

TL;DR

This paper reviews how machine learning and computational methods are transforming consumer data analysis by handling large-scale, complex data sources like text, images, and audio to better understand consumer behavior.

Contribution

It introduces recent data sources and machine learning techniques for analyzing consumer behavior at scale, bridging computational social science and marketing research.

Findings

Machine learning enables effective processing of multi-modal consumer data.

New data sources like crowdfunding and crowdsourcing reveal emerging behavioral patterns.

Computational methods address challenges of big data in consumer research.

Abstract

The digital revolution has led to the digitization of human behavior, creating unprecedented opportunities to understand observable actions on an unmatched scale. Emerging phenomena such as crowdfunding and crowdsourcing have further illuminated consumer behavior while also introducing new behavioral patterns. However, the sheer volume and complexity of this data present significant challenges for marketing researchers and practitioners. Traditional methods used to analyze consumer data fall short in handling the breadth, precision, and scale of emerging data sources. To address this, computational methods have been developed to manage the "big data" associated with consumer behavior, which typically includes structured data, textual data, audial data, and visual data. These methods, particularly machine learning, allow for effective parsing and processing of multi-faceted data. Given these recent developments, this review article seeks to familiarize researchers and practitioners with new data sources and analysis techniques for studying consumer behavior at scale. It serves as an introduction to the application of computational social science in understanding and leveraging publicly available consumer data.