A Crowdfunding Campaign Success Prediction Scheme Using Machine Learning

Using internet platforms, crowdfunding is a popular method used by entrepreneurs and innovators to raise funds for theirinitiatives. The success of a crowdfunding campaign depends on many variables, making it challenging to predict how it will play out. Prior research on predicting crowdfunding success had a number of drawbacks, such as the use of incomplete, noisy, and skewed datasets, a deficiency of features, imprecise prediction outcomes, and incompatibility with various campaign kinds or platforms. After examining a sizable dataset, the proposed machine learning (ML)-based crowdfunding campaign success prediction technique presented in this paper is able to predict crowdfunding campaign success results with higher accuracy. Several ML techniques, such as decision trees, logistic regression, support vector machines, K-nearest neighbors, and random forests, were employed in this study to identify the most effective prediction model. Random forest classifiers showed better results than other ML models, according to the resultant data, with 96.42% accuracy, 94.87% precision, 97.36% recall, and 96.98% F1-score. The proposed crowdfunding campaign success prediction model beats existing works by at least 3.1% in accuracy, 2.6% in precision, and 2% in recall score value, according to the performance comparison result. Additionally, this paper offers a mobile application for crowdfunding campaign assistance that has features like police reporting capabilities, donation options, suggestion options, and crowdfunding success prediction. According to the customer feedback results, over 80% of the evaluators were satisfied with the features of our program. Received: 14 September 2025 | Revised: 12 November 2025 | Accepted: 11 December 2025 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement The data that support this work are available upon reasonable request to the corresponding author. Author Contribution Statement Suzab-Ud-Doula: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Visualization. Mahfuzulhoq Chowdhury: Conceptualization, Methodology, Validation, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration.