In the field of financial market investment, quantitative investment is rapidly developing, which uses mathematical models and computer algorithms to mine the laws of financial data to provide a basis for investment decisions, effectively reduce human interference, and achieve efficient portfolio management and risk control. This study focuses on the application of machine learning algorithms in quantitative investment strategies, which collect and pre-process financial data from multiple sources, construct strategies using multiple algorithms, and generate results through historical backtesting and simulated trading. The study finds that different algorithmic models have large differences in returns, deep learning algorithms have significant advantages in processing time series data, and traditional machine learning algorithms are limited by linear assumptions and other limitations. The market environment has a key impact on algorithmic strategies, and the performance of algorithms varies in different market environments. This study points out the limitations of the research, provides suggestions on strategy selection for quantitative investment practice, and looks forward to future research directions, aiming to improve the effectiveness and adaptability of quantitative investment strategies and help financial market investment.