A well-liked method of funding startups, social causes and creative projects is through crowdfunding, where large numbers of people contribute relatively small amounts of money to fund a project. Nonetheless, conventional crowdfunding sites have issues of fraud, excessive charges, insufficient transparency and loss of investor control. Blockchain technology provides a better solution by allowing secure, transparent, and automatic handling of funds without middlemen. This paper provides an exploration of the ways through which blockchain technology enhances crowdfunding leveraging smart contract arrangements, decentralized organizations such as DAOs, milestone fund disbursement mechanisms, and tokenization. The interpretation of recent studies is made to indicate that such features as inability to change record-keeping and direct execution contribute to trust and security. System scalability, legal framework, and smart contract risk are discussed as challenges at present. It implements fraud detection using different AI Models such as SVM, Random Forest, XGBoost, (SMOTE + TabTransformer) and achieves accuracy of 92%, 93%, 93% and 95% respectively.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Blockchain-Enabled Decentralized Crowdfunding with Smart Contracts and AI based Fraud Detection Mechanisms

  • Ratnesh Kumar Choudhary,
  • Jai Patel,
  • Ansh Mishra,
  • Shivam Badade,
  • Rashi Yadav,
  • Ranjeet Shahu

摘要

A well-liked method of funding startups, social causes and creative projects is through crowdfunding, where large numbers of people contribute relatively small amounts of money to fund a project. Nonetheless, conventional crowdfunding sites have issues of fraud, excessive charges, insufficient transparency and loss of investor control. Blockchain technology provides a better solution by allowing secure, transparent, and automatic handling of funds without middlemen. This paper provides an exploration of the ways through which blockchain technology enhances crowdfunding leveraging smart contract arrangements, decentralized organizations such as DAOs, milestone fund disbursement mechanisms, and tokenization. The interpretation of recent studies is made to indicate that such features as inability to change record-keeping and direct execution contribute to trust and security. System scalability, legal framework, and smart contract risk are discussed as challenges at present. It implements fraud detection using different AI Models such as SVM, Random Forest, XGBoost, (SMOTE + TabTransformer) and achieves accuracy of 92%, 93%, 93% and 95% respectively.