Research on the Application of Improved Decision Tree Algorithm based on Information Entropy in the Financial Management of Colleges and Universities

In the era of information technology, work relies on technology to generate a huge amount data and information. Among them, financial universities is growing exponentially, manual method organizing extracting key can no longer meet requirements management universities. Taking higher education institutions as an example, it difficult grasp progress budget execution with frequent complicated daily expenditure income problems, then execute correct decisions in management. The study uses entropy decision basis tree institutions. value generated management, prediction accuracy tree. metric calculation introduced obtain well gain rate predict likelihood problematic events. validates performance improved dataset that achieves maximum 95% experiment. With accuracy, for university system, warning established link between current month's mechanism analyzed, finally two common algorithms, (Iterative Dichotomiser3，ID3) ID3 (Classification regression tree，CART) CART, are compared algorithm proposed study. mean square error sum squared metrics used conclude has better performance. By improving existing algorithm, proposes model based entropy, which aims help makers quickly accurately distill relevant make large more rational