EAGER: An Integrated Predictive Modeling Framework for Crowdfunding Environments

National Science Foundation Award # 1646881

Project Duration: 08/15/16-07/31/18

Principal Investigator:

Chandan Reddy

Graduate Students:

Vineeth Rakesh

Ping Wang

Tian Shi

Collaborators:

Jaegul Choo, Korea University

Project Summary:

The research aims to study data analytics tools for improving crowdfunding project success rate. Crowdfunding provides seed capital for start-up companies, creating job opportunities and reviving lost business ventures. In spite of the widespread popularity and innovativeness in the concept of crowdfunding, however, many projects are still not able to succeed. A deeper understanding of the factors affecting investment decisions will not only give better success rate to the future projects but will also provide appropriate guidelines for project creators who will be seeking funding. The crowdfunding domain poses several new challenges from the data analytics perspective due to the heterogeneous, complex and dynamic nature of the data associated with project campaigns. This project develops a systematic data-driven approach to resolve these challenges by utilizing vast amounts of historical data which can be leveraged to accurately predict the success of crowdfunding projects. Though the proposed methods are primarily developed in the context of crowdfunding, they are applicable to various other forms of social data that will be collected in other disciplines such as social science, engineering, and finance.

This project develops an integrated predictive modeling framework to solve some of the complex underlying problems related to bringing success to crowdfunding based projects. Existing approaches in data analytics for classification and regression cannot tackle this project success prediction problem since the goal is to estimate the time for a project to reach its success. The research team develops a unified probabilistic prediction framework which simultaneously integrates classification and regression together. In addition, a novel iterative imputation mechanism, which calibrates the time to project success, is proposed for reducing the bias in the model estimators. This project can demonstrate the power of data analytics in delivering better insights about various categories of real-world projects by not only accurately estimating the chances of being successful but also quantitatively assessing the factors that are responsible for bringing success in crowdfunding environments. The progress of the project and the research findings are disseminated via the project website (http://dmkd.cs.vt.edu/projects/crowdfunding/).

Journal Articles and Conference Papers:

• Vineeth Rakesh, Niranjan Jadhav, Alexander Kotov, and Chandan K. Reddy, "Probabilistic Social Sequential Model for Tour Recommendation", In Proceedings of the ACM International Conference on Web Search and Data Mining (WSDM), Cambridge, UK, February 2017. [pdf]
  

• Mahtab J. Fard, Ping Wang, Sanjay Chawla, and Chandan K. Reddy, "A Bayesian Perspective on Early Stage Event Prediction in Longitudinal Data", IEEE Transactions on Knowledge and Data Engineering (TKDE), Vol.28, No.12, pp.3126-3139, December 2016. [pdf]
  

• Vineeth Rakesh, Wang-Chien Lee, and Chandan K. Reddy, "Probabilistic Group Recommendation Model for Crowdfunding Domains", In Proceedings of the ACM International Conference on Web Search and Data Mining (WSDM), San Francisco, CA, February 2016. [pdf]
  

• Yan Li, Vineeth Rakesh, and Chandan K. Reddy, "Project Success Prediction in Crowdfunding Environments", In Proceedings of the ACM International Conference on Web Search and Data Mining (WSDM), San Francisco, CA, February 2016. [pdf]
  

• Vineeth Rakesh, Jaegul Choo, and Chandan K Reddy, "Project Recommendation using Heterogeneous Traits in Crowdfunding", In Proceedings of International AAAI Conference on Web and Social Media (ICWSM), Oxford, UK, May 2015. [pdf]
  

Software Download:

Will be made available soon.

dataset.

Point of Contact: Chandan Reddy : reddy (at) cs (dot) vt (dot) edu

Date of Last Update: 10/31/16