Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/250261 
Year of Publication: 
2021
Citation: 
[Journal:] Istanbul Business Research (IBR) [ISSN:] 2630-5488 [Volume:] 50 [Issue:] 2 [Publisher:] Istanbul University Press [Place:] Istanbul [Year:] 2021 [Pages:] 255-274
Publisher: 
Istanbul University Press, Istanbul
Abstract: 
Kickstarter is one of the popular crowdfunding platforms used to implement business ideas on the web. The success of crowdfunding projects such as Kickstarter is realized with future financial support. However, there is no platform where users can get decision support before presenting their projects to supporters. To solve this problem, a platform where users can test their projects is required. Within this scope, a business intelligence model that works on the web has been developed by combining business analytics and machine learning methods. The data used for business analytics has been brought to a state that can provide inferences through visualization, reporting and query processes. Within the scope of machine learning, various algorithms were applied for success classification and the best results were given by 91% Random Forest, 85% Decision Tree, 84% K-Nearest Neighbors (KNN) algorithms. F1-Score, Recall, Precision, Mean Squared Error (MSE), Kappa and AUC values were analyzed to determine the most successful models. Thus, Kickstarter users will be able to see their shortcomings and have a prediction about success before presenting their projects to their backers.
Subjects: 
Machine Learning
Business Analytics
Decision Support System
Business Intelligence
Persistent Identifier of the first edition: 
Creative Commons License: 
cc-by-nc Logo
Document Type: 
Article

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.