Mining Frequent Patterns in Uncertain and Relational Data Streams using the Landmark Windows

Todays, in many modern applications, we search for frequent and repeating patterns in the analyzed data sets. In this search, we look for patterns that frequently appear in data set and mark them as frequent patterns to enable users to make decisions based on these discoveries. Most algorithms presented in the context of data stream mining and frequent pattern detection, work either on uncertain data, or use the sliding window model to assess data streams. Sliding window model uses a fixed-size window to only maintain the most recently inserted data and ignores all previous data (or those that are out of its window). Many real-world applications however require maintaining all inserted or obtained data. Therefore, the question arises that whether other window models can be used to find frequent patterns in dynamic streams of uncertain data.In this paper, we used landmark window model and time-fading model to answer that question. The method presented in the form of proposed algorithm, which uses the idea of landmark window model to find frequent patterns in the relational and uncertain data streams, shows a better performance in finding functional dependencies than other methods in this field. Another advantage of this method compared with other methods is that it shows tuples that do not follow a single dependency. This feature can be used to detect inconsistent data in a data set.