Response models based on bagging neural networks

Abstract

Identifying customers who are likely to respond to a product offering is an important issue in direct marketing.Response models are typically built from historical purchase data. A popular method of choice, logistic regression, is easy to understand and build, but limited in that the model is linear in parameters. Neural networks are nonlinear and have been found to improve predictive accuracies for a variety of business applications. Neural networks have not always demonstrated clear supremacy over traditional statistics competitors, largely because of over-fitting and instability. Combining multiple networksmay alleviate these problems. A systematic method of combining neural networks is proposed, namely bagging or bootstrap aggregating, whereby overfitted multiple neural networks are trained with bootstrap replicas of the original data set and then averaged. We built response models using a publicly available DMEF data set with three methods: bagging neural networks, single neural networks, and conventional logistic regression. The proposed method not only improved but also stabilized the prediction accuracies over the other two.

Focusing on non-respondents: Response modeling with novelty detectors

“…A few of them have been applied to response modeling such as under-sampling (Ha, Cho, & MacLachlan, 2005), ensemble (Yu & Cho, 2006) and cost-modifying methods (Shin & Cho, 2006), among which the under-sampling method has been very popular. We apply the under-sampling method to LR.…”

“…While selecting or extracting relevant variables is very important, it is not our main concern. Malthouse (2001) extracted 17 out of the 91 input variables for this dataset, and Ha et al (2005) used 15 among them, removing two variables whose variations are negligible. In this paper, these 15 variables were used as input variables that are listed in Table 1.…”

Low-Cost Implementation of a Named Entity Recognition System for Voice-Activated Human-Appliance Interfaces in a Smart Home

“…However, they still need human annotation efforts for constructing the initial training corpus. To resolve this problem, we propose a semi-supervised NER system using active learning [16] based on bagging (bootstrap aggregating) [17] with distant supervision [18]. Unlike existing ML-systems, our system does not require a substantial amount of NE tagged training corpus and instead, only requires a NE dictionary that contains NEs and their classes.…”

Constructing response model using ensemble based on feature subset selection

“…The response rate was 9.4%. For practical reasons such as computational cost and imbalanced data distribution, only a subset of DMEF4 customers was selected and used for experiments based on 'weighted dollar amount spent' as was defined in (Ha, Cho, & MacLachlan, 2005):…”

“…However, complexities of human behavior rendered these inherently linear models have a lot to be desired. Recently, nonlinear and complex machine learning approaches were proposed such as neural networks (Bentz & Merunkay, 2000;Bounds & Ross, 1997;Ha, Cho, & MacLachlan, 2005;Moutinho, Curry, Davies, & Rita, 1994;Kim & Street, 2004;Zahavi & Levin, 1997a), decision trees (Haughton & Oulabi, 1997) and support vector machines (Cheung, Kwok, Law, & Tsui, 2003). They are known to be more flexible so that any degree of complexity can be learned or handled.…”