- Our Research
Where Insight Meets Intelligence
Advancing the frontiers of AI-driven research in finance, healthcare, education, and other priority sectors to power smarter decisions and sustainable innovation.
- Cost Predictive System
- Startup Risk Predictive System
- Cancer Risk Predictive System
Research from Gauge
December 31, 2024
AI/ML Research: Predicting the Cost of Consulting Projects
by Abiodun Gabriel Ajanaku
Consulting project cost estimation is complex due to varying project scopes, limited data, and human bias. This work proposes a machine learning-based system trained on a synthetic dataset of 2000 projects—derived from 80 real-world consulting cases using 17 expert-defined features. Five models were tested, with XGBoost outperforming others (MAPE 5%, R² = 99%). The most predictive feature was person-months (MI score: 89.7%). This system shows that ML can deliver high-accuracy cost predictions and suggests future work in integrating dynamic, real-time factors for enhanced generalization.
June 08, 2025
AI/ML Research: Startup Risk and Investment Readiness Predictve System
by Abiodun Gabriel Ajanaku
This study introduces an AI-driven system for evaluating startup risk and investment readiness in Africa, addressing the continent’s early-stage funding gap. Using a synthesized dataset of 10,000 startups modeled on real cases and covering five key risk dimensions—financial, operational, compliance, technology, and strategic—the system applies machine learning to assess venture viability. After narrowing to 19 influential features, five models were benchmarked, with Random Forest achieving the best performance (F1 score: 0.98) and optimized for high recall to avoid misclassifying high-risk startups. The model offers a scalable, data-driven tool for investors and ecosystem stakeholders to make informed, low-risk investment decisions in emerging markets.
July 22, 2024
AI/ML Research: Predicting the Risk of Cancer in Adult
by Abiodun Gabriel Ajanaku, Olufemi Isiaq
This study presents a machine learning-based system for classifying cancer risk in adults using survey data. Drawing on 580 participant records with 58 features, six ML models were developed, with Random Forest achieving the highest performance (AUC: 0.93, Recall: 93%). The final model identified 10 key predictors—including lifestyle habits, environmental exposures, and comorbidities—to classify individuals into low- or high-risk cancer groups. By integrating often-overlooked comorbid conditions, the system offers a more holistic and accurate approach to cancer risk prediction, aiding early detection and targeted awareness efforts.
July 22, 2024
Research & Thought Leadership
by Abiodun Gabriel Ajanaku, Olufemi Isiaq
Stay ahead of industry trends and establish your organization as a thought leader. Our research services combine rigorous academic methodology with practical business insights, creating authoritative content that positions you at the forefront of AI innovation.
From Real-World Data to Real-World Impact
We conduct research in AI systems designed to understand, model, and simulate complex real-world dynamics—empowering data-driven solutions across finance, healthcare, education, and beyond.
We're doing research in AI systems that can understand and simulate the world and its dynamics.
AI/ML Research: Startup Risk and Investment Readiness Predictive System
Artificial Intelligence for Startup Risk and Investment Readiness Assessment: A Machine Learning Model from the African Innovation Ecosystem
AI/ML Research: Startup Risk and Investment Readiness Predictive System
Artificial Intelligence for Startup Risk and Investment Readiness Assessment: A Machine Learning Model from the African Innovation Ecosystem
AI/ML Research: Startup Risk and Investment Readiness Predictive System
Artificial Intelligence for Startup Risk and Investment Readiness Assessment: A Machine Learning Model from the African Innovation Ecosystem