PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
April 16, 2026Open Access

Early-Stage Startup Failure Prediction Using Machine Learning Models

View Full Paper
Ask AI
Bookmark
Share

Authors

DRDr. K. RangaswamyBRB. Sai Sreekar Reddy

Discussion

Loading...

Member takes

Overview

This research evaluates startup durability in a volatile landscape, suggesting improved risk assessment through machine learning.

Key Points

  • The aim is to predict startup success or failure using machine learning models based on various features.
  • Analyzed 923 startup profiles with 15 features across financial, operational, and geographical dimensions.
  • Utilized several machine learning architectures including Logistic Regression, Random Forests, and Gradient Boosting.
  • Developed a soft-voting ensemble method to enhance prediction accuracy.
  • Implemented synthetic indicators for capital efficiency and milestone tracking.
  • Achieved a classification accuracy of 81.3% using the tuned Random Forest model.
  • F1-score was found to be 0.865, indicating strong predictive capability.
  • Demonstrated that algorithmic feature engineering enhances reliability in risk assessment.

Cite This Study

Rangaswamy et al. (2026) studied this question.

synapsesocial.com/papers/69e07e582f7e8953b7cbf53fhttps://doi.org/10.5281/zenodo.19564116
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Predicting Startup Survival Using Machine Learning2026
  2. 2A FRAMEWORK TO PREDICTING STARTUP SUCCESS GROWTH WITH MULTIPLE ALGORITHMS2025
  3. 3Advanced Predictive Models for the Startup Ecosystem Using Machine Learning Algorithms2024
  4. 4Data-Driven Framework for Classification and Management of Start-Up Risk for High Investment Returns2024 · 3 citations
  5. 5A Leakage-Controlled, Calibration-First Evaluation of Machine Learning Models for Startup-Outcome Prediction: Evidence from Crunchbase2026