inglese
Crowdfunding has become a powerful mechanism for collective financing, offering a global reach and versatile applications across various sectors. It challenges traditional funding sources like bank loans, venture capital, and private equity. This work delves into the complex dynamics of crowdfunding platforms, focusing on investor behaviour and investment patterns within equity and lending campaigns. By employing advanced machine learning techniques, such as XGBoost and LSTM networks, I have developed predictive models that analyse real-time and historical data to accurately forecast the success or failure of crowdfunding campaigns. A distinctive aspect of this research is the introduction of a "pre and post-launch" analysis frame-work, which examines the factors influencing campaign success both before the launch (e.g., marketing strategies, initial investor interest) and after (e.g., ongoing investor engagement and funding momentum). This dual-phase perspective addresses a criti- cal gap in existing studies, offering a more comprehensive understanding of campaign dynamics. To further enhance crowdfunding analysis tools, I introduce two novel datasets, one for equity crowdfunding and another for lending, tailored to capture the unique tem- poral and behavioural patterns within Italian crowdfunding platforms. Additionally, my approach moves beyond traditional binary success metrics, proposing innovative measures that better capture campaign outcomes, such as funding ratios, overfunding levels, and long-term visibility metrics. The insights gained from this study can significantly enhance crowdfunding strate- gies, improving project selection, pre-launch preparations, and post-launch promo- tional tactics on platforms. By refining decision-making processes and offering forward-looking guidance to investors, our computational model empowers both campaign cre-ators and platform administrators. Ultimately, this research contributes to increasing the overall efficacy and sustainability of crowdfunding as a financing tool
Data Science approaches to support the design of crowdfunding campaigns
PORRO, ROSA
2025
Abstract
inglese| File | Dimensione | Formato | |
|---|---|---|---|
|
Thesis_Rosa_Porro_signed_signed_signed.pdf
accesso aperto
Licenza:
Tutti i diritti riservati
Dimensione
1.64 MB
Formato
Adobe PDF
|
1.64 MB | Adobe PDF | Visualizza/Apri |
|
Thesis_Rosa_Porro_signed_signed_signed_1.pdf
accesso aperto
Licenza:
Tutti i diritti riservati
Dimensione
1.64 MB
Formato
Adobe PDF
|
1.64 MB | Adobe PDF | Visualizza/Apri |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/358266
URN:NBN:IT:UNIBA-358266