A Machine Learning-Based Suitable Startup Recommendation Scheme and Investor Assistance Mobile Application

Abstract

Investing in startups is a complex decision-making process that requires identifying suitable companies and matching them with compatible investors. The existing works did not present any intelligent startup suitability prediction for funding and investor assistance mobile applications by taking into account growth rate, annual profit, rating, and reputation factors. This paper initiates a machine learning-based recommendation system for predicting suitable startups and matching them with appropriate investors. The proposed system leverages a comprehensive dataset of startup and investor characteristics, including market segment, geographical region, and city. In comparison with singular value decomposition and principal component analysis schemes, our results show that Term
Frequency-Inverse Document Frequency scheme is selected for startup recommendation due to its high recall and precision value. Our investor assistance mobile application offers features like login, apply for funds, best funder recommendation, best entrepreneur selection, and user rating features. Our application evaluation results indicated that more than 60 percent of users are satisfied with the feasibility of the proposed mobile application.

Received: 23 February 2024 | Revised: 9 April 2024 | Accepted: 14 May 2024

Conflicts of Interest 

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

The dataset used in this paper is available on the Crunchbase website.

Author Contribution Statement

Asfak Asif: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization. Mahfuzulhoq Chowdhury: Conceptualization, Methodology, Investigation, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration.