An Effective Topic Modeling Strategies for Recommender Systems in Crowdfunding Platforms

Suresh Subramanian

Data and Metadata, 2024, vol. 3, .349

Abstract: Capitalists come up with creative and innovative concepts, but a lack of finance limits their untapped economic potential. There are several channels that new entrepreneurs may use and take advantage of to attract money and other financial resources when beginning a firm thanks to current technology, which has drastically altered the way business is done on a broad scale. An entrepreneur uses the Internet to promote his concept to potential backers through crowdfunding. Online crowdfunding has labored to develop several advanced platforms that may serve as an interface to the fundraising process for a certain concept or project. Typically, the owner of the concept explores the market and does extensive research through a variety of channels, with the Internet assisting in moving ahead and making the idea actual. In truth, the owner of the concept frequently suffers obstacles and financial issues, therefore crowdsourcing helps to alleviate these issues. In this study, machine learning methods were used to train the system on the given data, beginning with the theme, followed by the blurb, which is the topic description, and finally by the topic category. Latent Dirichlet Allocation (LDA) and Latent Semantic Analysis (LSA) were employed as machine learning approaches to accomplish the goal. This study employs a variety of text classification algorithms, including Support Vector Machine (SVM), EXtreme Gradient Boosting (XG), K-Nearest Neighbours (KNN), and Random Forest (RF), to propose and forecast subject categories. Each algorithm performed differently in terms of precision, predictability, positive rate, and model correctness. SVM was the highest performance measuremen.

Date: 2024
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:dbk:datame:v:3:y:2024:i::p:.349:id:1056294dm2024349

DOI: 10.56294/dm2024.349

Access Statistics for this article

More articles in Data and Metadata from AG Editor
Bibliographic data for series maintained by Javier Gonzalez-Argote ().