Extracting comprehensible patterns from Venezuelan assets by means of annotated Traffic Lights Pannel

In this work the relationship between Venezuelan assets and the general Index of the Venezuelan Stock Exchange is analyzed by means of data mining methods. In particular, clustering followed by the Traffic Lights Panel visualization of cluster prototypes is used to understand the meaning of the discovered patterns. Also, associations between Venezuelan index and Dow Johnes (NY) or BOVESPA (Brazil) show the association with international context. The work confirms a well-known fact, that is the self-behaviour or Venezuelan stock market, often disconnected from international behaviour. Also, intrinsic uncertainty associated with the class prototypes is introduced into the visualization through annotated-TLP and more reliable or stable patterns can be distinguished from more variable, or volatile.