KDDLab research produced a number of free opensource data mining software, such as:
- Trajectory Pattern Miner: Data mining tool for the extraction of spatio-temporal frequent patterns from GPS traces of moving objects. Sourceforge
- Never Walk Alone: Exploiting spatial location uncertainty for ensuring anonymity in moving objects databases. Website
- Temporally Annotated Sequences: Data mining tool for the extraction of sequential patterns with typical transition times between elements of the sequences.
- Daedalus: Knowledge Discovery support environment for data mining queries on mobility data.
- Athena: Semantic enrichment environment for trajectories and pattern intepretation and understanding. athena has been integrated with Daedalus thus provinding a complete tool for data mining querying and reasoning.
- T-Cluster: density-based clustering algorithm to discover set of similar trajectories, according to a
repertoire of trajectory similarity functions
- WhereNext: tool for the prediction of the next location of a moving object, which uses the previous movements of all objects in a certain area to learn a classifier.
- GERM: the Graph Evolution Rule Miner: given a set of temporal snapshots of the same single graph, this miner finds all the evolution patterns more frequent than a given threshold minfreq, and no bigger than a given threshold maxsize, from which we can compute rules of evolution, together with their support and confidence.
- GAMP, the Graph Antimonotone-Monotone Pruner: given a dataset of
graphs, a threshold of frequency minfreq and a conjunction of monotone and antimonotone constraints, this preprocessor performs a loop of pruning and returns a dataset of graphs where each graph satisfies the monotone constraints, and each edge is frequent and satisfies the antimonotone constraints.
- M-Atlas M-Atlas is a mobility querying and data mining system centered onto the concept of trajectory. Besides the mechanisms for storing and querying trajectory data, M-Atlas has mechanisms for mining trajectory patterns and models that, in turn, can be stored and queried. Website
- SMA Sequence Mining Automata: a new techniques for mining frequent sequences under regular expressions. Website
- BF-P2KA BF-2PkA is an algorithm for anonymizing sequence data. It transforms a sequence dataset in a k-anonymous version while preserving the sequential pattern mining analysis. Sourceforge