With the rapid penetration of big data technology

the demand for data analysis capabilities in various industries has gradually changed. Probability and statistics serve as the core course for cultivating data analysis thinking

yet the traditional teaching model lags behind in terms of content

methods

and evaluation systems. Therefore

this paper

based on the new requirements for talents' capabilities in the big data era

analyzes the existing problems in the current teaching of probability and statistics courses

such as the over-emphasis on theoretical derivation in content

the lack of interactivity in teaching methods

and the singleness of assessment methods. From three dimensions—updating teaching content

innovating teaching methods

and reconstructing the assessment system—it proposes reform strategies

including integrating big data cases from multiple fields

strengthening the teaching of Python/R tools

promoting the project-driven model

and constructing process-oriented evaluation. These strategies aim to enhance the adaptability of the course to big data applications

cultivate students' core competitiveness in solving practical problems

and provide references for talent cultivation under the background of emerging engineering and emerging liberal arts.