Financial data analysis plays a crucial role in investment decision-making, but investors must also be wary of the issues that may arise. To address issues such as information lag, financial data manipulation, and distorted financial indicators. This paper employs hierarchical clustering and principal component analysis methods to conduct a diagnostic study of the financial condition of HK Company. The study primarily involves selecting financial condition diagnostic indicators, performing dimensionality reduction on the diagnostic indicators, and extracting the principal components of the diagnostic indicators. Subsequently, a comprehensive evaluation function is constructed based on the contribution rates of each principal component. This evaluation function enables the determination of the company’s current financial condition, providing reliable data support for investment decisions. A collaborative filtering algorithm based on weighted triads is proposed as an investment decision-support model to provide investment decision schemes for investors. Experimental analysis indicates that the proposed model outperforms the benchmark method in estimating user preferences with greater accuracy. It also addresses the data sparsity issue where most results are zero when calculating the similarity between investment products using traditional collaborative filtering recommendation algorithms.
