Welcome to MAT-Tools’s documentation!

---

mat-tools

MAT-tools: Python Framework for Multiple Aspect Trajectory Data Mining

The present project is an umbrella project for MAT-tools Framework. The following projects are created as tools to support the user in the data mining of multiple aspect trajectories, specifically for classification, summarization, clustering, extracting and visualizing the movelets (the parts of the trajectory that better discriminate a class label). It integrates into a unique platform the fragmented approaches available for multiple aspects trajectories and in general for multidimensional sequence classification into a unique web-based and python library system.

• Short Demonstration Video

Datasets Repository

A pre-processed repository of multiple aspect datasets is available with the MAT-Tools Framework.

Core Package Dependencies

Package	Description
mat-model	Dependency package for model classes and data structures for MAT.
mat-data	Preprocessing and dataset generation tools for MAT data

Data Mining Tools

Package	Description
mat-view	Web-interface and library tools for MAT and movelets visualization
mat-similarity	MAT similarity methods.
mat-classification	MAT classification methods (movelet-based or otherwise).
mat-clustering	MAT clustering methods (similarity-based and others).
automatize	Older package for MAT classification and Web-interface for MAT analysis

---

Dependency Packages:

• MAT-data
• MAT-model

Data Mining Tools:

• MAT-similarity
• MAT-classification
• MAT-clustering
• MAT-view