Feature subset selection for predicting the success of crowdfunding project campaigns

Statistics from crowdfunding platforms show that a small percent of crowdfunding projects succeed in securing funds. This makes project creators eager to know the probability of success of their campaign and the features that contribute to its success before launching it on crowdfunding platforms. The existing literature focuses on examining success probability using the entire list of identified projects features. For situations for which project creators have limited resources to invest on the required project features, the list suggested by previous researchers is somewhat large and gives a small success probability. A minimal number of features that predict success with a higher probability can benefit project creators by providing them with insight and guidance in investing their limited resources. This paper presents a metaheuristic whale optimization algorithm (WOA) in the crowdfunding context to perform a complete search of a subset of features that have a high success contribution power. Experiments were conducted using WOA with the K-Nearest Neighbor (KNN) classifier on a Kickstarter dataset. Our approach obtains a subset of 9 features that predict the success of project campaigns with an accuracy (F-score) of 90.28% (90.11%), which is an increase (F-score) of 22.23% (21.61%) than when a complete set of features is used. The findings of this study contribute knowledge to various crowdfunding stakeholders, as they will provide new insights regarding a subset of essential features th4at influence the success of project campaigns with high accuracy.

[1]  Chih-Ping Wei,et al.  Will Your Project Get the Green Light? Predicting the Success of Crowdfunding Campaigns , 2015, PACIS.

[2]  Paul Belleflamme,et al.  Crowdfunding: Tapping the Right Crowd , 2013, SSRN Electronic Journal.

[3]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[4]  José García,et al.  Putting Continuous Metaheuristics to Work in Binary Search Spaces , 2017, Complex..

[5]  Ferat Sahin,et al.  A survey on feature selection methods , 2014, Comput. Electr. Eng..

[6]  Isabelle Guyon,et al.  An Introduction to Variable and Feature Selection , 2003, J. Mach. Learn. Res..

[7]  Fahad Sarfaraz Ahmad,et al.  Predicting crowdfunding success with optimally weighted random forests , 2017, 2017 International Conference on Infocom Technologies and Unmanned Systems (Trends and Future Directions) (ICTUS).

[8]  Weiguo Fan,et al.  Project description and crowdfunding success: an exploratory study , 2018, Inf. Syst. Frontiers.

[9]  Hiroshi Motoda,et al.  Feature Extraction, Construction and Selection: A Data Mining Perspective , 1998 .

[10]  Li-Yeh Chuang,et al.  Improved binary particle swarm optimization using catfish effect for feature selection , 2011, Expert Syst. Appl..

[11]  El-Ghazali Talbi,et al.  A Taxonomy of Hybrid Metaheuristics , 2002, J. Heuristics.

[12]  Hossam Faris,et al.  Simultaneous Feature Selection and Support Vector Machine Optimization Using the Grasshopper Optimization Algorithm , 2018, Cognitive Computation.

[13]  Gilbert Laporte,et al.  Metaheuristics: A bibliography , 1996, Ann. Oper. Res..

[14]  Gaël Varoquaux,et al.  The NumPy Array: A Structure for Efficient Numerical Computation , 2011, Computing in Science & Engineering.

[15]  El-Ghazali Talbi,et al.  Metaheuristics - From Design to Implementation , 2009 .

[16]  B. Bayus,et al.  Crowdfunding Creative Ideas: The Dynamics of Project Backers , 2018 .

[17]  Hiroshi Motoda,et al.  Feature Extraction, Construction and Selection , 1998 .

[18]  Ivelin Elenchev,et al.  Forecasting the Success Rate of Reward Based Crowdfunding Projects , 2017, Managing Global Transitions.

[19]  Ethan Mollick.  The Dynamics of Crowdfunding: An Exploratory Study , 2014 .

[20]  Vincent Etter,et al.  Launch hard or go home!: predicting the success of kickstarter campaigns , 2013, COSN '13.

[21]  Hiroshi Motoda,et al.  Computational Methods of Feature Selection , 2022 .

[22]  Aboul Ella Hassanien,et al.  Binary grey wolf optimization approaches for feature selection , 2016, Neurocomputing.

[23]  Yan Li,et al.  Project Success Prediction in Crowdfunding Environments , 2016, WSDM.

[24]  GunopulosDimitrios,et al.  Locally Adaptive Metric Nearest-Neighbor Classification , 2002 .

[25]  Maoguo Gong,et al.  Influence maximization in social networks based on discrete particle swarm optimization , 2016, Inf. Sci..

[26]  Aboul Ella Hassanien,et al.  Binary ant lion approaches for feature selection , 2016, Neurocomputing.

[27]  Andrew Lewis,et al.  The Whale Optimization Algorithm , 2016, Adv. Eng. Softw..

[28]  Wes McKinney,et al.  Data Structures for Statistical Computing in Python , 2010, SciPy.

[29]  Ahmed Bouridane,et al.  Simultaneous feature selection and feature weighting using Hybrid Tabu Search/K-nearest neighbor classifier , 2007, Pattern Recognit. Lett..

[30]  Harmeet Kaur,et al.  Effect of Social Media Connectivity on Success of Crowdfunding Campaigns , 2017, ITQM.

[31]  Majdi M. Mafarja,et al.  Hybrid Whale Optimization Algorithm with simulated annealing for feature selection , 2017, Neurocomputing.

[32]  Dimitrios Gunopulos,et al.  Locally Adaptive Metric Nearest-Neighbor Classification , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[33]  Namita Srivastava,et al.  A fuzzy based feature selection from independent component subspace for machine learning classification of microarray data , 2016, Genomics data.

[34]  Elizabeth Gerber,et al.  Crowdfunding support tools: predicting success & failure , 2013, CHI Extended Abstracts.

[35]  Peter Harrington,et al.  Machine Learning in Action , 2012 .