Incremental Temporal Association Rules Algorithm And Its Application

In recent years, with the development of computer technology and extensive application of database technology, the industry has accumulated a mass of different forms of data storage, from the data found valuable information or knowledge and using it for decision-making is a very difficult task, therefore data mining came into being. At the same time, in real life, we find that a large number of data sets with either explicit or implicit characteristics of the time, the temporal data mining researches have gradually become a new hot spot.Temporal data are more time-sensitive. In other words, the data will change with time change. Since the original has been excavated or knowledge of the model may no longer be effective, they need to re-excavation, but the original work has been wasted. Therefore, we propose a time-bound under the incremental mining association rules algorithm, as new data to generate incremental update association rules set, as far as possible to deal only with new data and make full use of the excavation of the original rules.The algorithm is based mainly at the state event model and serial model and Apriori principle, fast update algorithm. In practical application of great significance, not only on the stock data, but also on data such as banks, supermarkets data, meteorological data, we can analysis and research these temporal data.In this paper, we use incremental temporal association rules algorithm for stock data analysis with the stock k-line theory. Finally, we arrive at a number of interesting association rules in order to guide investors to choose the investment strategy.