A survey and benchmarking study of multitreatment uplift modeling

Abstract

Uplift modeling is an instrument used to estimate the change in outcome due to a treatment at the individual entity level. Uplift models assist decision-makers in optimally allocating scarce resources. This allows the selection of the subset of entities for which the effect of a treatment will be largest and, as such, the maximization of the overall returns. The literature on uplift modeling mostly focuses on queries concerning the effect of a single treatment and rarely considers situations where more than one treatment alternative is utilized. This article surveys the current literature on multitreatment uplift modeling and proposes two novel techniques: the naive uplift approach and the multitreatment modified outcome approach. Moreover, a benchmarking experiment is performed to contrast the performances of different multitreatment uplift modeling techniques across eight data sets from various domains. We verify and, if needed, correct the imbalance among the pretreatment characteristics of the treatment groups by means of optimal propensity score matching, which ensures a correct interpretation of the estimated uplift. Conventional and recently proposed evaluation metrics are adapted to the multitreatment scenario to assess performance. None of the evaluated techniques consistently outperforms other techniques. Hence, it is concluded that performance largely depends on the context and problem characteristics. The newly proposed techniques are found to offer similar performances compared to state-of-the-art approaches.

Heterogeneous treatment effect analysis based on machine‐learning methodology

“…Overall, both approaches are based on the random forest model and aim to address similar HTE questions (e.g., both can be used for survival analysis 40–43 ). Although our study focused on HTE analysis with a single treatment, it is worth noting that the uplift model and causal forest are applicable in circumstances with multiple possible treatments 44–46 …”

“…Although our study focused on HTE analysis with a single treatment, it is worth noting that the uplift model and causal forest are applicable in circumstances with multiple possible treatments. 44 , 45 , 46…”

Generalized Causal Tree for Uplift Modeling

“…Following Olaya et al (2020), we convert continuous outcome to a binary outcome in AOD for a fair comparison. Detailed descriptions of the two datasets are provided in Section D.…”

“…The expected response of a uplift model is then defined as R " 1 N ř N i"1 R i , which is an unbiased estimator of Ery i |T " vpx i qs. Following Olaya et al (2020), we first apply propensity scoring matching to debias the non-experimental data. Then we fit models using 5-fold cross-validation, and the results from each round are averaged to obtain the overall performance.…”

Uplift modeling VS conventional predictive model: A reliable machine learning model to solve employee turnover

“…There are various prescriptive analytics models for HR that have been studied in the literature. For instance, an HR recruitment model [18]- [20] and employee turnover uplift (ETU) model [21], [22]. Furthermore, Floris et al [23] suggest that uplift modeling may be widely applied, for instance in marketing, personalized medicine, and political election.…”