The main concern in information-rich systems is to efficiently navigate and access desired information. Traversing a file system using long pathnames is cumbersome and requires the user to accurately remember where each file can be found. Semantic file systems help in finding a file under various contexts (called as tags) which act like directories. However, the user still has to traverse these paths to reach the files. It is left up to the user to create and manage an efficient system of tags. Since all data items and relations are stored in a database structure, association rule learning can be used to form useful relations between various tags and files. Using various algorithms, we can create associations (links) between data sets that can help the user traverse related data without the burden of long pathnames. Popular data mining algorithms can be readily adapted to a semantic file system's database. Thus utilizing associations, a semantic file system can offer a more efficient and contextual way to search, store and organize data.