The integration of machine learning and data-driven methods into railway condition monitoring has shown significant potential for improving safety and operational efficiency. A critical aspect of this advancement is the analysis of contact force data measurements obtained from instrumented wheelsets, which provide real-time insights into wheel-rail interaction dynamics. By leveraging high-resolution sensor data, machine learning algorithms can detect anomalies, for an optimization of maintenance schedules, reducing downtime and costs. This paper explores the application of data-driven techniques to process and interpret contact force measurements, enabling early fault detection and enhanced decision-making. Instrumented wheelsets serve as a key enabler, offering precise and reliable data streams that feed into classifiers and predictive models. Space-frequency techniques such as Short Time Fourier Transform are employed to parse the data and obtain useful features for analysis. The study highlights the challenges and opportunities in deploying instrumented wheelsets, space-frequency analysis techniques and machine learning for railway condition monitoring. Case studies demonstrate the potential of these approaches in improving the accuracy of wear and defect detection while supporting proactive maintenance strategies. The findings underscore the potential of combining advanced analytics with instrumented wheelset technology to enhance railway asset management and operational reliability.