Predictive Modeling With Big Data: Is Bigger Really Better?

By Enric Junqué de Fortuny, David Martens, Foster J. Provost. 2013

183 citations recorded by the source corpus.

Abstract With the increasingly widespread collection and processing of “big data,” there is natural interest in using these data assets to improve decision making. One of the best understood ways to use data to improve decision making is via predictive analytics. An important, open question is: to what extent do larger data actually lead to better predictive models? In this article we empirically demonstrate that when predictive models are built from sparse, fine-grained data—such as data on low-level human behavior—we continue to see marginal increases in predictive performance even to very large scale. The empirical results are based on data drawn from nine different predictive modeling applications, from book reviews to banking transactions. This study provides a clear illustration that larger data indeed can be more valuable assets for predictive analytics. This implies that institutions with larger data assets—plus the skill to take advantage of them—potentially can obtain substantial competitive advantage over institutions without such access or skill. Moreover, the results suggest that it is worthwhile for companies with access to such fine-grained data, in the context of a key predictive task, to gather both more data instances and more possible data features. As an additional contribution, we introduce an implementation of the multivariate Bernoulli Naïve Bayes algorithm that can scale to massive, sparse data.

Subfields: Artificial Intelligence, Information Systems, Marketing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Marketing

Citations in the corpus

Explaining Classification Models Built on High-Dimensional Sparse Data (2016, background): “However, with fine-grained behavior data it has been shown that feature selection can lead to substantially reduced predictive performance (Junqué de Fortuny et al., 2013).”
The Impact of Big Data on Firm Performance: An Empirical Investigation (2019, unknown): “De Fortuny et al. (2013) discuss the performance of a variant of Naive Bayesian classification algorithm on a number of data sets from online content providers as well as a bank.”
Metaverse for Healthcare: A Survey on Potential Applications, Challenges and Future Directions (2022, background): “providing better sentiment and behavioral analytics [150], predictive support [151] and fraud detection [152], [153].”
Dimensionality Reduction via Matrix Factorization for Predictive Modeling from Large, Sparse Behavioral Data (2015, background): “The established value of such massive, high-dimensional data (Junqu´e de Fortuny et al. 2013) leads to more interest in utilizing it.”
Mining Massive Fine-Grained Behavior Data to Improve Predictive Analytics (2016, background): “The intuitive explanation for this is that people can only make a finite number of such choices in a limited amount of time (Junqué de Fortuny et al. 2013).”
Big Data Research in Information Systems: Toward an Inclusive Research Agenda (2016, background): “Many other recent studies also suggest that using more data instances and variables can improve predictive capabilities (Junque de Fortuny et al., 2013).”
A comparison of instance-level counterfactual explanation algorithms for behavioral and textual data: SEDC, LIME-C and SHAP-C (2019, background): “The proliferation of big data architectures has resulted in predictive modeling applications having an increasingly large impact on business and society [16].”
State of the art review: the data revolution in critical care (2015, background): “The most intuitive argument (that more data from which to learn cannot be worse, so must be better) is true: There have been empirical demonstrations that predictive models built from sparse, fine-grained data see marginal gains in predictive performance even to massive scale [40].”