A Comparison and Interpretation of Machine Learning Algorithm for the Prediction of Online Purchase Conversion

Lee, Jung-won; Jung, Okkyung; Lee, Yunhye; Kim, Ohsung; Park, Cheol

doi:10.3390/jtaer16050083

Cited by 32 publications

(18 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In cases where the paper showed several methods and results, the model with the best performance was chosen. The lowest response rate in the previously used datasets was 2.29% (conversion rate) [14], while the highest was 27.42% [7], which is significantly higher than the response rate in this study.…”

Section: Literature Reviewcontrasting

confidence: 80%

“…Several recent papers treat this issue regarding online direct marketing campaigns and overall online purchase prediction using web log data. For instance, Lee et al [14] explored machine learning models and potential effective data sampling methods for predicting online consumer behaviour for the visitors of a Google Merchandise Store. The authors found that the eXtreme Gradient Boosting (XGB) algorithm is most effective for predicting purchase conversion of online consumers, while oversampling with the SMOTE algorithm was shown to be the best method to solve the data imbalance issue.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Response models for direct online campaigns involving web metrics are increasingly relevant nowadays, as indicated by newer research studies [14][15][16]. The problem of class imbalance is especially pronounced because the response rate is lower due to the large number of user accesses without response (website browsing, which does not end with a purchase).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Customer Response Model in Direct Marketing: Solving the Problem of Unbalanced Dataset with a Balanced Support Vector Machine

Rogić

Kašćelan

Bach

2022

JTAER

View full text Add to dashboard Cite

Customer response models have gained popularity due to their ability to significantly improve the likelihood of targeting the customers most likely to buy a product or a service. These models are built using databases of previous customers’ buying decisions. However, a smaller number of customers in these databases often bought the product or service than those who did not do so, resulting in unbalanced datasets. This problem is especially significant for online marketing campaigns when the class imbalance emerges due to many website sessions. Unbalanced datasets pose a specific challenge in data-mining modelling due to the inability of most of the algorithms to capture the characteristics of the classes that are unrepresented in the dataset. This paper proposes an approach based on a combination of random undersampling and Support Vector Machine (SVM) classification applied to the unbalanced dataset to create a Balanced SVM (B-SVM) data pre-processor resulting in a dataset that is analysed with several classifiers. The experiments indicate that using the B-SVM strategy combined with classification methods increases the base models’ predictive performance, indicating that the B-SVM approach efficiently pre-processes the data, correcting noise and class imbalance. Hence, companies may use the B-SVM approach to more efficiently select customers more likely to respond to a campaign.

show abstract

Section: Literature Reviewcontrasting

confidence: 80%

Section: Literature Reviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Customer Response Model in Direct Marketing: Solving the Problem of Unbalanced Dataset with a Balanced Support Vector Machine

Rogić

Kašćelan

Bach

2022

JTAER

View full text Add to dashboard Cite

show abstract

“…The application of machine learning algorithms to the marketing field has raised much attention due to the availability of big data along with the complex marketing environment, which is increasingly difficult to predict [34]. Ma and Sun [35] provided a systematical overview of marketing research with machine learning methods.…”

Section: Discussionmentioning

confidence: 99%

Dynamic Marketing Resource Allocation with Two-Stage Decisions

Zhang

Liao

et al. 2022

JTAER

View full text Add to dashboard Cite

In the precision marketing of a new product, it is a challenge to allocate limited resources to the target customer groups with different characteristics. We presented a framework using the distance-based algorithm, K-nearest neighbors, and support vector machine to capture customers’ preferences toward promotion channels. Additionally, online learning programming was combined with machine learning strategies to fit a dynamic environment, evaluating its performance through a parsimonious model of minimum regret. A resource optimization model was proposed using classification results as input. In particular, we collected data from an institution that provides financial credit products to capital-constrained small businesses. Our sample contained 525,919 customers who will be introduced to a new product. By simulating different scenarios between resources and demand, we showed an up to 22.42% increase in the number of expected borrowers when KNN was performed with an optimal resource allocation strategy. Our results also show that KNN is the most stable method to perform classification and that the distance-based algorithm has the most efficient adoption with online learning.

show abstract

“…In recent years, with the further development of artificial intelligence, we have also witnessed and been surprised at how AI has become part of our popular culture. Especially since 1997, when International Business Machine Corporation (IBM)'s Deep Blue defeated Garry Kasparov, the world chess champion, people have gained a deeper understanding of AI [6][7][8]. Nowadays, intelligent personal assistants (IPAs) have become popular and common among premiumtier smartphones.…”

Section: Introductionmentioning

confidence: 99%

Analyzing Determinants for Adoption of Intelligent Personal Assistant: An Empirical Study

Sun

2021

Applied Sciences

View full text Add to dashboard Cite

With the accelerating advances in artificial intelligence (AI) technologies, the rising demand for the intelligent personal assistant is expected to rapidly increase in the following decades. As customer loyalty has become a critical element in gaining a competitive advantage and many marketers pay great attention to it, this research aims to explore the factors that directly affect customer loyalty toward the intelligent personal assistant. The main motivation is to investigate and identify the direct effects that variables such as customer innovativeness and perceived value have on customer loyalty. Additionally, both product knowledge and corporate image will be examined as moderators. The results obtained demonstrate that perceived value has a positive influence on customer loyalty. Furthermore, product knowledge moderates the influence of functional and hedonic innovativeness on customer satisfaction. The study’s main contributions provide a better understanding of customer loyalty. This research aims to be viewed as a specific guide to provide effective practices that companies should implement to help them reallocate resources appropriately and enhance customer value creation. The results indicate that the higher the level of customer perceived value, the better the customer loyalty will be. The findings represent that corporate image has a positive moderating effect on the influence of customer innovativeness on perceived value. This study draws on the outcomes of administrative suggestions and recommends a few observational strategies to improve the monetary administration in the intelligent personal assistant industry.

show abstract

A Comparison and Interpretation of Machine Learning Algorithm for the Prediction of Online Purchase Conversion

Cited by 32 publications

References 67 publications

Customer Response Model in Direct Marketing: Solving the Problem of Unbalanced Dataset with a Balanced Support Vector Machine

Customer Response Model in Direct Marketing: Solving the Problem of Unbalanced Dataset with a Balanced Support Vector Machine

Dynamic Marketing Resource Allocation with Two-Stage Decisions

Analyzing Determinants for Adoption of Intelligent Personal Assistant: An Empirical Study

Contact Info

Product

Resources

About