application of machine learning in marketing
DESCRIPTION
The study of methods or algorithms meant to understand the underlying patterns in data and generate predictions based on these patterns is known as machine learning (ML). In marketing, academic research has typically focused on causal inference. Read More with Us: https://bit.ly/3taGLqJ Why Statswork? Plagiarism Free | Unlimited Support | Prompt Turnaround Times | Subject Matter Expertise | Experienced Bio-statisticians & Statisticians | Statistics across Methodologies | Wide Range of Tools & Technologies Supports | Tutoring Services | 24/7 Email Support | Recommended by Universities Contact Us: Website: www.statswork.com Email: [email protected] United Kingdom: +44 1618184707 India: +91 4446313550 WhatsApp: +91 8754467066TRANSCRIPT
Application ofmachine learningin marketingAn Academic presentation by
Dr. Nancy Agnes, Head, Technical Operations, Statswork Group www.statswork.comEmail: [email protected]
Outline
TODAY'S DISCUSSION
INTRODUCTION
APPROACHES
TYPES OF MODELS
SUPPORT VECTOR MACHINES
The study of methods or algorithms meant to understand the underlyingpatterns in data and generate predictions based on these patterns is knownas machine learning (ML).
In marketing, academic research has typically focused on causal inference.
The requirement to generate counterfactual predictions drives the focus oncausation.
Will rising advertising spending, for example, improve demand? To answerthis question, you'll need an unbiased assessment of the influence ofadvertising on demand.
Introduction
Marketing techniques, on the other hand, rely on the ability to make correctforecasts.
For example, which customers to target, which product configurations acustomer is most likely to pick, which form of a banner ad will produce the mostclicks, and what rivals' market shares and actions are likely to be.
All of these are issues of prediction.
These issues do not necessitate causality; instead, great out-of-sampleprediction accuracy is required.
Machine learning technology can help with these difficulties.
Both in terms of their emphasis and the features they supply, machine learningapproaches differ from econometric methods.
To begin, machine learning approaches are concerned with achieving the best out-of-sample predictions, whereas causal econometric methods are concerned withproducing the best-unbiased estimators.
As a result, methods designed for causal inference frequently fail to performeffectively when making out-of-sample predictions.
The best-unbiased estimator does not always offer the most incredible out-of-sampleprediction, as we shall explain below, and in some instances, a biassed estimatorperforms better for out-of-sample data.
Approaches
Second, machine learning techniques are built to function in circumstances wherewe don't have an a priori understanding of how the data's results were created.
This feature of ML differs from econometric techniques, which are used toevaluate a specific causal hypothesis.
Third, unlike many empirical marketing strategies, machine learning algorithmscan handle many variables and determine which ones should be kept and whichshould be eliminated.
Finally, with ML approaches, scalability is a significant concern and strategies likefeature selection and efficient optimization aid in achieving scale and efficiency.
Because many of these algorithms must operate in real-time, scalability isbecoming increasingly crucial for marketers.
Second, machine learning techniques are built to function in circumstances wherewe don't have an a priori understanding of how the data's results were created.
This feature of ML differs from econometric techniques, which are used to evaluatea specific causal hypothesis.
Third, unlike many empirical marketing strategies, machine learning algorithms canhandle many variables and determine which ones should be kept and which shouldbe eliminated.
Finally, with ML approaches, scalability is a significant concern and strategies likefeature selection and efficient optimization aid in achieving scale and efficiency.
Types of models
Because many of these algorithms must operate in real-time, scalability is becomingincreasingly crucial for marketers.
Support Vector Machines is a prominent classification algorithm in the last20 years, with applications in image recognition, text mining, and illnessdetection, thanks to its stability and capacity to handle massive, high-dimensional data.
Cui and Curry (2005) brought it to marketing and introduced SVM theory andapplications.
They also compare SVM's predictive performance to that of the multinomiallogit model on simulated choice data, demonstrating that SVM outperformsthe multinomial logit model, especially when data is noisy and products havea large number of characteristics (i.e., high dimensionality).
Support Vector Machines
They also discovered that SVM beats the multinomial logit model by a substantialmargin when predicting choices from more extensive choice sets.
Although both techniques' predictive performance decreases as the options setgrow, the reduction is significantly steeper for multinomial logit than for SVMbecause the first-choice prediction task becomes more complex.
The latent-class SVM model, which permits the inclusion of latent variables insideSVM, is another extension.
By elaborating on the convex–concave technique used to estimate latent-classSVM while incorporating respondent heterogeneity, Liu and Dzyabura (2016)create an algorithm for predicting multi-taste consumer preferences.
They demonstrate that their model's prediction outperforms single-tastebenchmarks.
