Summary 01: Project 01
Machine learning in budget forecasting for corporate finance: A conceptual model
for improving financial planning
The article explores the application of machine learning (ML) in budget forecasting for corporate finance,
proposing a conceptual model to enhance accuracy and efficiency in financial planning. It broadly refers
to the use of historical financial dataset. It addresses the limitations of traditional forecasting methods:
linear regression and moving averages, by leveraging advanced ML techniques like LSTM networks,
ARIMA, and ensemble methods. It proves potentially improving accuracy by 10-30% or more, depending
on data quality and model implementation. For instance: LSTM networks in time series forecasting
typically achieve 85-95% accuracy in well-structured datasets. Ensemble methods like Random Forests
and Gradient Boosting Machines further enhance robustness and reduce errors.
Summary 01: Project 02
Optimizing Upsell and Cross-Sell Strategies Using Reinforcement Learning and
Collaborative Filtering Algorithms
The article explores the integration of reinforcement learning and collaborative filtering algorithms to
optimize upsell and cross-sell strategies. Using a retail dataset, the hybrid model dynamically adapts
recommendations based on real-time customer interactions, addressing challenges like the cold start
problem and evolving user behavior. The dataset includes transactional data. The approach significantly
improves personalization, customer satisfaction, and revenue growth, outperforming traditional methods.
While specific accuracy metrics are not provided, the model is expected to enhance conversion rates and
average order value (AOV) by 10-30% or more, outperforming traditional methods.
Citations
1. Olamijuwon, J. and Zouo, S.J.C., 2024. Machine learning in budget forecasting for corporate finance:
A conceptual model for improving financial planning.
2. Singh, V., Chopra, S., Singh, P. and Joshi, P., 2020. Optimizing AI-Driven Upsell and Cross-Sell
Strategies Using Reinforcement Learning and Collaborative Filtering Algorithms. Journal of AI ML
Research, 9(4).