DSpace Repository

Business Revenue Prediction: A Case Study of Restaurant Franchises

Show simple item record

dc.contributor.advisor Guha, Sumanta
dc.contributor.author Agarwal, Shilpa
dc.contributor.other Esichaikul, Vatcharaporn
dc.contributor.other Anutariya, Chutiporn
dc.date.accessioned 2016-05-12T08:34:18Z
dc.date.available 2016-05-12T08:34:18Z
dc.date.issued 2016-05
dc.identifier.issn AIT
dc.identifier.uri http://www.cs.ait.ac.th/xmlui/handle/123456789/820
dc.description thirty eight pages en_US
dc.description.abstract This research study examined the problem of predicting the revenue of a proposed new business. In particular, this study addressed the case of a restaurant franchise and attempted to develop efficient and reliable techniques to predict revenue at a proposed location based on historical data of existing franchise locations. Various techniques including Support Vector Machine, Random Forest, Logistic Regression and Linear Regression were evaluated with the R Programming tool. This study adapted an approach which included data preprocessing, selections of best suitable mode, model validation and performance evaluation and bias correction of the outputs. In data preprocessing, certain characteristics of the data set were examined and potential fixes with different techniques were provided. Evaluation of various models indicated that Random Forest shows the best performance. In reality lowering the RMSE is not the only consideration when creating a model, interpretability and speed also matter. The results obtained here can further be improved using various data mining techniques, ensembles techniques or applying bias correction methods. en_US
dc.description.sponsorship AIT Fellowship en_US
dc.language.iso en_US en_US
dc.publisher AIT en_US
dc.subject SVM, Random Forest, Cross fold validation en_US
dc.title Business Revenue Prediction: A Case Study of Restaurant Franchises en_US
dc.type Research report en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search DSpace


Advanced Search

Browse

My Account