Software

Bayesia announces the release of BayesiaLab 4.5, the software for efficient Data Mining with Bayesian Networks.

The main new features of this release are:

• Filtered/Censored States to rigorously take into account the variables that are contingent on context (usage/configuration).
• Network Comparison in terms of arcs, edges and V-Structures.
• Arc Confidence Analysis and Targeted Cross-Validation based on the JackKnife and k-folds methods respectively. Possibility to automatically extract the Bayesian network based on user-defined constraints on the arc frequencies.
• Fixed Probabilities: the tool for entering evidences as probability distributions allows now to fix these distributions. The likelihoods are recomputed after each new pieces of evidence gathered on the other nodes so that the fixed probability distributions of the nodes remain the same as initially defined.
• Evidence Scenarios to associate a set of evidences (hard evidences, likelihood distributions and fixed probability distributions) to a Bayesian network. These scenarios can then be used for interactive inference, interactive updating, and all the batch processes.

You will find on our website the newsletter that describes more in details all the 70 new features of BayesiaLab 4.5, and new presentations related to Filtered/Censored States, Probabilistic Structural Equations, Driver analysis, and Customer Segmentation.

Please, feel free to download BayesiaLab 4.5 for a new 30 days evaluation period.