Abstract:
Because the rise of crowdfunding, entrepreneurs decrease seeking help from traditionally financial institutions, but began to get help on the Internet. Now, more than 15 ...Show MoreMetadata
Abstract:
Because the rise of crowdfunding, entrepreneurs decrease seeking help from traditionally financial institutions, but began to get help on the Internet. Now, more than 15 million people involved and the amount of funds raised exceeded 3.9 billion US dollars. Although crowdfunding provides a new fundraising channel for entrepreneurs who need to raise funds, success in reaching the target amount is a big challenge. How to increase the success rate of fundraising projects is the most concern of all fundraisers. Most of the current researches aimed to explore the relation between the founders and the success of the project. Relatively few works focus on the impact of the description and the wording of the project to predict the success rate of fundraising. Therefore, this study will collect real fundraising projects from Kickstarter, and analyze the text description content of these projects. The feature selection method, Support Vector Machines Recursive Feature Elimination (SVM-RFE), has been employed to find key words that may affect the success of the project. Then, we'll use selected keywords to build a prediction model by utilizing Support Vector Machines (SVM) to help emerging entrepreneurs or anyone who needs to raise funds can have a higher chance of successful fundraising.
Published in: 2019 IEEE 6th International Conference on Industrial Engineering and Applications (ICIEA)
Date of Conference: 12-15 April 2019
Date Added to IEEE Xplore: 16 May 2019
ISBN Information: