Along with the fast increase of data in many emerging 
applications, a great deal of research effort has been 
elaborated upon the data mining area to discover interesting 
but unknown knowledge from large databases. Mining of 
association rules is usually stated as: “Given a database of 
transactions, one would like to discover important elationships 
among items such that the presence of some items in a transaction 
will imply the presence of other items in the same transaction.” 
Association rule mining has been known to be useful in areas like 
decision support, market data analysis, recommendation systems, 
and Web data management, to name a few. 

In this talk, I will share with the audience some of my recent 
works that address practically interesting issues on the mining 
of association rules. First, we shall consider the association 
mining that retrieves top k itemsets in the presence of memory
 constraint. Explicitly, we shall take into consideration 
the available memory space for discovering frequent itemsets,
 while allowing users of only providing the desired number of 
frequent itemsets (instead of the minimum support count). 
Second, we shall explore a new general model of association 
mining in a temporal database, where the exhibition periods of
items are allowed to be different from one to another. 
Temporal association rules derived will enable users to observe 
short-term but interesting patterns that would not be found when 
the whole range of the database is evaluated together as in most
prior works. Third, we examine the mining in a data stream 
environment and devise a regression-based algorithm for 
association mining in that environment. In a data stream 
environment, the volume of data is usually too huge to be stored 
on permanent devices or to be scanned for more than once. Hence, 
both approximation and adaptability are key ingredients for 
performing mining tasks over rapid data streams. Finally, we 
address the issue of hardware-enhanced mining. A hash-based and 
pipelined architecture for hardware-enhanced association 
mining is presented. In view of the explosion of data amount in 
emerging mining applications, the hardware-enhanced mining is 
considered a potentially important research direction to 
explore for future data mining tasks.