Application And Research Of Data Mining Technology In College Teaching Evaluation

Data mining extracts from large, and noise, vague and random data which is implied in themselves and it is potentially useful the process of effective information and knowledge. Data mining technology can solve the current database systems face a "data rich and lack of knowledge" this issue. Association rule mining is the most important and most mature one of the research methods and techniques that can find the services meaningful connection between the hidden and rules.The purpose of this study is data mining techniques combined with the teaching evaluation, evaluation of data from a large extract useful information hidden, so as to educational administration departments to provide decision support basis for improving teaching quality.In this paper, the classical association rule mining algorithm Apriori algorithm is described and analyzed in detail based on their need for multiple scans the transaction database and the set of possible candidates for the two large performance bottleneck problem, a new improved Apriori algorithm-IA algorithm is proposed that based on adjacency table. IA algorithm is introduced in "The length transaction" is a new concept and the use of adjacency table to store transaction database. Ck in the IA algorithm for support counting step for each candidate, by reducing the scanning of k-frequent itemsets generated invalid transaction methods to improve efficiency. By experiment, when the length of the transaction in the database of different especially when the difference between the larger, IA algorithm is more efficient than Apriori algorithm.Finally, we discuss IA algorithm in the teaching evaluation of specific applications. The school's teaching evaluation data are analyzed and processed, dig out the relationship between the indicators, on the basis of the proposals put forward views on improving teaching quality.