AgenaRisk 10 最新更新 - 贝叶斯网络设计软件-上海卡贝信息技术有限公司

We are pleased to announce the release of Version 10 of AgenaRisk, which represents a revolutionary breakthrough in Bayesian network modelling, exploiting state-of-the-art research in AI and smart data to quantify risk and reduce uncertainty.

While Version 10 retains all of the power and ease of use of its predecessors, its new features provide functionality unmatched in any other tools, and it builds on extensions to its revolutionary algorithm for accurately modelling ‘hybrid’ models – those that contain both discrete and (unconstrained) continuous variables. In particular, AgenaRisk 10 is the first to provide the following in-demand functionality for such hybrid models:

Solve ‘influence diagrams’: you can now add (multiple) decision and utility nodes to models and calculate the optimal strategy to maximize overall utility or minimize overall risk. The solution is also presented as an easy to understand decision tree.
Calculate of Value of Information (VoI): The new VoI functionality automatically calculates the ‘value’ of uncertain variables in terms of how much it is ‘worth’ to find out more about them. It is used to prioritise the parts of a decision model where additional information is important for decision making.
‘Smart’ learning from data: Version 10 extends the smart learning that was introduced for discrete variables in the previous version of AgenaRisk to continuous variables and mixtures of continuous and discrete. The ‘smart’ learning enables automatic learning of probability tables and expressions from a spreadsheet file – with or without missing data - incorporating as much or as little expert judgement as required.
We are launching our AgenaRisk Developer product designed for use by programmers to makes it easier to develop a tailored implementation using AgenaRisk or to integrate AgenaRisk into an existing application.

Version 10 includes new statistical distribution functions, such as BetaPERT, more example models in our model library, and the inference algorithms have been revamped so that complex models run faster and more accurately.

A new licensing scheme in Version 10 ensures a much simpler, smoother setup process and more flexible licensing. As usual license holders, with a current annual subscription, are entitled to a free upgrade.

Remember to join the AgenaRisk LinkedIn User Group if you have not already done so. As well as answering user questions, this is where we now post news, tips and tricks, report on case studies and link to the latest research.