BayesiaLab 5.0: New Features

• Data
  • New format for arc constraint dictionary
  • Disambiguation of the names of classes, nodes and states in the dictionaries
  • BOM taken into account when exporting dictionaries
  • State virtual number dictionary
  • Local structural coefficient dictionary
  • Rearranging dictionaries
  • Database report
  • Choice of unnecessary columns when associating database
  • Row identifier
  • Information about the other types of columns
• Network
  • Costs greater or equal to 1
  • Select nodes with missing values
  • Select nodes with assessments
• Learning
  • K-Means Clustering
  • New parameters for Taboo learning
  • New parameters for Taboo Order learning
  • Structure equivalent example number for each learning algorithm
  • Keeping learning parameters
• Inference
  • Observed mean/value
  • Fixing current probabilities
  • Relaxing probabilities
  • Causal inference
  • New database save dialogs
  • Selecting database rows with identifier
• Analysis
  • Variation editor for analysis
  • Upgrade of target dynamic profile
  • Target optimization tree
  • Target interpretation tree
  • Kolmogorov-Smirnov test in network global performance
  • Extract database in global performance
  • Fix reference in influence analysis
  • Arc's mutual information
  • Filtered values taken into account in mutual information
  • Colored borders in influence analysis
  • Translucent nodes skipped in analyzes
  • Filtered states skipped in target report
  • Translucent arcs skipped in relationship analysis
  • Hidden variable discovery
  • Target optimization
  • Probability analysis wrt the target state
• Monitors
  • New menu Monitor
  • Log-likelihood in monitor information panel
  • Sorting monitors
  • Fitting the monitor's size
  • Scientific notation
  • Tooltip with extended values int the information panel of the monitors
  • Fix and relax probabilities
  • Display properties saved
• Interface
  • Three state stop button
  • Colored names in reports
  • Colored line plot
  • Layer intersection
  • Import/Export in property editors
  • Sortable tables in property editors
• Knowledge Elicitation
  • Assessments
  • Experts
  • Online assessment editing
  • Assessment tools
  • Assessment sensitivity analysis
  • Assessment report
• Tools
  • Exporting assessments
  • Shuffle on data in cross-validation
  • Joint comparison
  • Structural coefficient modified in cross-validation
  • Database stratification in cross-validation
  • Structural coefficient analysis
  • Data perturbation
  • Multi-quadrant analysis
• Settings
  • Variable clustering settings
  • Managing instance reusability
  • Scientific notation
  • Perspectives
• Help
  • Analysis of the use of BayesiaLab
• Shortcuts
  • Changing target
  • Exclusion of nodes
  • Total effects
  • Zoom
• Ergonomics
  • Windows, Linux and others
  • Mac OS X
  • Setup 64 bits for Windows
  • Error e-mail
• Miscellanous
  • Analyst edition

Data

• Arc
• Node
• State

It contains two parts. The first part is a global summary analysis of the database:

It shows the number of variables, the number of variables with missing values and the number of variables with continuous values associated.

The global database is analyzed and the number of examples is indicated. If the database has stratification on a node, it is also indicated with the corresponding probability distribution. The sum of weights and the normalization factor of the weights are displayed if database has associated weights. The report indicates the number of missing values and if the database has row identifers.

If the database has data types associated, the learning and test databases are also analyzed. For each one, the example number, the sum of the weights and the number of missing values are shown.

The second part of the report details the content of the database for each variable:

For each discrete or continuous variable, the numbers of missing values and filtered values with their associated percentages are displayed if necessary. For the continuous variables, the report indicates if they have associated continuous values. It indicates also the minimum, the maximum and the mean of each continuous variable.

All this information is displayed for the global, learning and test databases if data types are associated.

The button Unmatched Columns displays all the columns in the database that are not in the network. The following dialog is displayed and allows the user to distribute or not the selected columns:

A new type of column is available when a database is imported or associated: setting a column of the database as row identifier allows defining an identifier for each row.

Identifiers can be of any type (string or numbers) but cannot contain missing values. However, unicity is not required. Two rows can have the same identifier. These identifiers will be saved with the database and kept in any derived database (generated by some analysis or tools).

This identifier can be used to select a line in the database that will be observed (during Interactive Inference, Interactive Updating or manual selection on the database). The current identifier is displayed in the status bar.

Network

The menu Edit>Select Nodes>Missing Values allows selecting the nodes having a percentage of missing values greater than a given threshold:

The menu Edit>Select Nodes>Assessments allows selecting the nodes having assessments.

Learning

The menu Learning>K-Means Clustering allows clustering data in an unsupervised way thanks to the k-means clustering algorithm, in order to find partitions of homogeneous elements. Originally used only on real variables, this algorithm is extended to use also discrete variables. Before running the k-means clustering, the user must select the variables on which he wants to apply the algorithm. These variables can be either continuous or discrete but not hidden (i.e. without associated data). Once selected, the clustering tries to regroup the data into a chosen number of clusters in which each observation belongs to the cluster with the nearest mean. It is similar to the expectation-maximization algorithm for mixtures of Gaussians in that they both attempt to find the centers of natural clusters in the data as well as in the iterative refinement approach employed by both algorithms.

The values of the selected nodes used by the clustering algorithm are computed as following:

1. If the node is continuous and there are continuous values in the database associated to this node, these continuous values are used.
2. If the node has values associated to its states, these values are used.
3. If the node is continuous, the center of each interval is used.
4. If the node is discrete with integer or real states, the values of the states are used.
5. And, if none of the previous cases is valid, integer values are given to each state starting from 0.

In all cases, the filtered values are skipped and the data are standardized.

If there are missing values in the database for the selected variables, the corresponding rows are skipped and not used for the clustering. The weights associated to the chosen rows are taken into account.

Once finished, a node Clusters is created with each state corresponding to each cluster found by the algorithm.

The data corresponding to the created node Clusters are added in the database. A missing value is put for the node in each row containing missing values for the selected variables.

The following dialog box allows entering the wanted number of clusters. Of course this number must be superior or equal to 2:

It is possible to define the size of the taboo list as well as the maximum numbers of parents and children allowed. If these options are not checked, they are not taken into account.

In addition to standard options, it is possible to maintain the current structure of the network to start learning.

It is possible to define the size of the taboo list and to maintain the current structure of the network to start learning.

For each learning algorithm there are startup options indicating that all arcs will be deleted before learning and allowing the user to define a Structure Equivalent Example Number, i.e. a virtual database with the indicated size representing the current structure:

For each parametrized learning algorithm, the chosen parameters are saved in the network and can be reused later in various tools.

Inference

Now, a target mean/value can be set as an evidence for a node.

When a node has values associated with states or is a continuous node or has numerical states, it is possible to choose a target mean/value for this node. An algorithm based on MinXEnt allows determining a probability distribution corresponding to this target mean/value, if this distribution exists. Of course, the indicated target value must be greater than or equal to the minimum value and less than or equal to the maximum value.

Once the target value entered, there are three options:

• No Fixing: the distribution found must be observed as likelihoods.
• Fix Mean: the indicated mean must be observed as fixed mean. When the mean is fixed, if an observation is done on another node, the convergence algorithm will automatically determine a new distribution in order to obtain the target mean, taking the other observations into account. If we store this evidence in the evidence scenario file, only the target mean will be stored. Fixing mean is also done in the evidence scenario files with the notation m{...}. You must note that fixing mean is only valid for the exact inference. If the approximate inference is used, fixing mean is considered like simply setting the likelihoods corresponding to the target mean: there is no more convergence algorithm.
• Fix Probabilities: the distribution found must be set as fixed probability distribution. Fixing probabilities is also done in the evidence scenario files with the notation p{...}. You must note that fixing probabilities is only valid for the exact inference. If the approximate inference is used, fixing probabilities is considered like simply setting the likelihoods corresponding to the target mean: there is no more convergence algorithm.

In addition to the standard exact inference, it is possible to make causal inference, ie to consider a set of nodes as causal nodes in the Bayesian network.

Until a node defined as causal has no evidence set, it behaves like a classical node. However, once evidence set, the node is actually causal: relationships with potential parents are deleted.

To consider a node as causal, just click on the menu Intervention in the monitor's contextual menu in validation mode. In this case, the monitor node becomes blue like a decision node. If evidence has already been set on this node, it will be automatically removed.

Once the mode of observation used is set to Intervention, the node is defined as causal and it is possible to set all possible kinds of evidence. Once evidence set on the node, the network is mutilated and the node takes the form of a square as the representation of a decision node.

In the next image, the node Smoking has been defined as causal by clicking on the menu Intervention, but it has not yet been observed. He behaves therefore as a normal node:

Once the node Smoking has evidence, it loses its relationship with its parent and is represented as a square:

To transform a causal node into classic, simply click on the menu Observation in the monitor's contextual menu. If an evidence has been set on this node, it will be automatically deleted.

Nodes defined as causal are taken into account in any analysis.

In the various database save dialogs (save database, batch tools with current database, etc.), the user can choose the part of the database (learning, test, all) he wants to save, if a data type is associated.

The number of selected nodes is displayed if any. In this case, the user can choose to save in the file only the selected nodes or all of them.

In addition, the user can choose to save the long name of the states and the associated continuous values.

In validation mode, if the database has row identifiers, a shift + right-click on the database icon displays a floating panel allowing the user to perform a search among the identifiers. The search is done thanks to a text field. The wildcards characters ? and * can be used. ? can replace one and only one character. * can replace none or several characters. The Case Sensitive option allows the search to respect the case. After pressing enter, the search is performed and the list of the results is displayed. The number of corresponding rows is displayed at the bottom of the panel. A click on a line sets the corresponding observations and the row identifier is displayed in the status bar:

This panel is also displayed in interactive inference and interactive updating with a database when the user performs a right-click on the textfield.

Analysis

Several analysis like Target Optimization, Target Optimization Tree, Target Interpretation Tree, Target Interpretation Tree and Target Dynamic Profile need to define positive and negative variations in order to apply them on the mean of each node. The following editor allows the user to edit these variations. This editor is accessible from the parameter panels of the concerned analysis. It is possible to associate with each node a negative and a positive variation in percentage. These variations will be saved with the network and will be available for each analysis.

A variation, positive or negative, is a positive real number between 0 and 100%. The default value is 10%.

The Import button allows importing a list of variations from a dictionary as in Associate Dictionary menu.
The syntax is:

The Export button allows exporting the list of negative and positive variations in a dictionary as in Export Dictionary menu.

The target dynamic profile has been dramatically upgraded. The modifications and additions are explained below:

Criterion Optimization:
In the criterion optimization area, the user can choose to take into account the costs associated with the nodes. In this case, the computed criterion is weighted by costs.

Search Method:
Four search methods for criterion optimization are available:

• Hard Evidences: Only hard evidences will be used for the optimization.
• Value/Mean Variations in %: The observation can be done either by fixing means, i.e. the convergence algorithm will find, at each new observation, a probability distribution to obtain the wanted mean, or by fixing the probabilities, i.e. at startup, the probability distributions corresponding to the wanted means are computed once for all and will not change anymore.
  The observed means will vary in percentage:
  • Of the initial means of the nodes: Each driver variable will see its initial mean vary according to the given negative and positive percentages. The computed mean is bounded by the variable's variation domain's limits.
  • Of the variation domains of the nodes: Each driver variable will see its initial mean vary according to the given negative and positive percentages of the variable's variation domain. The computed mean is bounded by the variable's variation domain's limits.
  • Of the margin progress of the nodes: Each driver variable will see its initial mean vary according to the percentages of the difference between the initial mean and the minimum of the domain and the difference between the initial mean and the maximum of the domain.
  Each node's mean is computed from the values associated to the states. If there is no associated value, if the node is continuous, its mean will be computed from the intervals, and if the node is discrete with integer or real states, the mean is computed from them. If there is no possibility to compute the mean, a default set of values from 0 to the number of states minus one is used.
  When an observed mean reach the limits of the variation domain of a node, the corresponding hard evidence is used instead.
  The percentages of the negative and positive variations of each node can be modified with the mean variation editor.

In each case, the variables' filtered states will never be used.

Stop Criterion:
The search is stopped when the joint probability of the network reaches zero. But this stop criterion can be modified by setting a maximum number to the evidence done and by modifying the minimum joint probability allowed.
It is also possible to use the automatic stop criterion that allows finding a good balance between search depth and usefulness of the levers.

This function allows searching for several policies in order to optimize the target node. Two options are available:

• Optimization of the specified target state
• Optimization of the target's mean

Its functionning is similar to the target dynamic profile except that instead of finding only the "optimal policy" it tests also alternative optimization policies.
The algorithm finds the best evidence on a node to optimize the target (state or mean). Once this node found, two branches in the tree are created:

• the left branch has the node observed and it stays observed in the subtree
• the right branch has no observation, and this node cannot be observed anymore in this subtree. It is called an alternate tree.

For each branch, the search is done again on the remaining nodes with the same rules until one of the stop criterions is obtained or no more node is available.

The filtered states are not taken into account in the search.

The following panel allows entering all the parameters:

You can select the target node. By default, the current target node is used. After that, you can choose to optimize the selected target state or the mean.

You can choose to maximize or minimize the criterion.
A checkbox allows taking the joint probability of the evidence into account during the search. In this case, we will optimize the a posteriori (probability of the target state knowing the evidence weighted by the occurrence probability of that evidence).
In the same way, the cost of evidence can be taken into account. The observation context is also taken into account during the analysis.

Several search methods are available:

• By hard evidence on nodes' states.
• By observation of the value/mean of the nodes. The observation can be done either by fixing means, i.e. the convergence algorithm will find, at each new observation, a probability distribution to obtain the wanted mean, or by fixing the probabilities, i.e. at startup, the probability distributions corresponding to the wanted means are computed once for all and will not change anymore.
  The observed means will vary in percentage:
  • Of the initial means of the nodes
  • Of the variation domains of the nodes
  • Of the margin progress of the nodes, i.e. the difference between the initial mean and the minimum of the domain and the difference between the initial mean and the maximum of the domain.
  The percentages of the negative and positive variations of each node can be modified with the mean variation editor.

The settings panel allows restricting the search with four different stop criterions that can be combined.

• You can specify the maximum number of nodes that can be observed in a branch of the tree.
• You can also specify the maximum number of alternate trees allowed. Alternate trees correspond to the nodes that are skipped in the right branches of the tree.
• The algorithm can stop also when an evidence reaches the minimum joint probability allowed.
• The automatic stop criterion is an heuristic test that allows us to find a good balance between the optimization due to an evidence and the impact on the joint probability.

The results will be associated as evidence scenario file. Each branch of the tree corresponds to a line of evidence in the evidence scenario file.

If an evidence file is already associated, you can choose to replace it or to append the examples at the end.

The results are displayed in a binary tree. Each node of the tree represents the node chosen by the algorithm.

The evidence context will be displayed at the top if necessary.

A node contains four parts:

• The name or comment of the target node. The used background is the node's color.
• The probability of the target state of the target node before setting the evidence if we optimize the target state or the mean of the target node before setting the evidence if we optimize the mean.
• The joint probability before setting the evidence.
• If the node is not a leaf, the name or comment of the chosen node that will be observed. The used background is the node's color.
  Otherwise, it shows the value of the optimization score with a white background.
  The last node of the "optimal branch" has a thick red border in order to find it quickly.

Two kinds of branches exist:

• The left branches on which the chosen nodes are observed. If hard evidence is used, the name of the observed state is displayed on the branch. If the mean is observed, the value of the mean of the chosen node is displayed. If this mean corresponds to the negative variation, it is displayed in red, otherwise, if it is the positive variation, it is displayed in blue.
• The right branches represent that the chosen node is not observed and skipped. It won't be used in the corresponding subtree.

The toolbar at the top of the window allows the user:

•  To zoom in the tree
•  To zoom out the tree
•  To go back to the initial zoom
•  To fit the tree to the window
•  To display the tree horizontally

The buttons at the bottom of the windows allow closing the window, saving the image, printing the tree and displaying the report.

The contextual menu allows:

• Displaying the comment instead of the name of the nodes
• Displaying the long names of the states
• Copying the tree as an image or as html. The html copy reproduces the vertical display as well as the horizontal one. The displayed colors are also used.

It is possible to perform a manual zoom by clicking and dragging the mouse to select the area that will be zoomed in. The shortcut ctrl + mousewheel performs zooming (and unzooming) on the tree centered on the cursor.

Moving the mouse over a node displays a tooltip containing the chain of evidence done until this node.

A double-click on a node closes its subtrees in the window. Double-click on it to display them again.

Depending on the options, the button Report displays an html report containing the list of the best chosen nodes in the whole tree with their best observation and the corresponding optimization score.

This function allows generating a tree that helps interpreting the target node. The algorithm optimizes the knowledge of the target by reducing its corresponding entropy (or binary entropy if we choose a specific state with the "Center on State" option). We are then using the (binary) mutual information to build the tree (that can be combined with the cost associated to the nodes).

The Adaptive Questionnaire is functionally close to that tree. If the user always choose the first variable, the sequence of the questionnaire is described in that tree.
Once the algorithm has found a node, n branches, corresponding to the n states of the node, are created in the tree if hard evidence is used, otherwise, two branches, correponding to the negative and positive variations of the mean, are created if the mean variations are used.
On each branch the observed state or mean is displayed.

For each branch, the search is done again on the remaining nodes with the same rules until one of the stop criterions is obtained or no more node is available.

The filtered states are not taken into account in the search.

The following panel allows entering all the parameters:

You can select the target node. By default, the current target node is used. After that, you can choose to center the search on a specific state or not.

Several search methods are available:

• By hard evidence on nodes' states.
• By observation of the value/mean of the nodes. The observation can be done either by fixing means, i.e. the convergence algorithm will find, at each new observation, a probability distribution to obtain the wanted mean, or by fixing the probabilities, i.e. at startup, the probability distributions corresponding to the wanted means are computed once for all and will not change anymore.
  The observed means will vary in percentage:
  • Of the initial means of the nodes
  • Of the variation domains of the nodes
  • Of the margin progress of the nodes, i.e. the difference between the initial mean and the minimum of the domain and the difference between the initial mean and the maximum of the domain.
  The percentages of the negative and positive variations of each node can be modified with the mean variation editor.

A checkbox allows taking the cost of the evidence into account during the search by dividing the score based on the mutual information.

The observation context is also taken into account during the analysis.

The settings panel allows restricting the search with three different stop criterions that can be combined.

• You can specify the maximum number of nodes that can be observed in a branch of the tree.
• The algorithm can stop also when an evidence reaches the minimum joint probability allowed.
• The automatic stop criterion is the value of the conditional mutual information (divided by the cost if the cost option is checked). The algorithm stops if this value is lower than the given parameter.

The results will be associated as evidence scenario file. Each branch of the tree corresponds to a line of evidence in the evidence scenario file.

If an evidence file is already associated, you can choose to replace it or to append the examples at the end.

The results are displayed in a binary tree if the mean variations are used or in a n-ary tree if hard evidence is used. Each node of the tree represents the node chosen by the algorithm.

The evidence context will be displayed at the top if necessary.

Nodes are very similar to monitors. A node contains four parts:

• The name or comment of the target node. The used background is the node's color.
• The value of the node with its delta and the joint probability with its delta before setting the evidence.
• The probability distribution of the target node before setting the evidence. Each state with the corresponding probability is displayed as in the monitor. If the search is centered on a state, this state is painted in light blue.
• If the node is not a leaf, the name or comment of the chosen node that will be observed. The used background is the node's color.
  The score based on the mutual information (divided by the cost if the cost option is checked) is displayed in percentage.

If hard evidence is used, the name of the observed state is displayed on the branch. If the mean is observed, the value of the mean of the chosen node is displayed. If this mean corresponds to the negative variation, it is displayed in red, otherwise, if it is the positive variation, it is displayed in blue.

The toolbar at the top of the window allows the user:

•  To zoom in the tree
•  To zoom out the tree
•  To go back to the initial zoom
•  To fit the tree to the window
•  To display the tree horizontally

The buttons at the bottom of the windows allow closing the window, saving the image and printing the tree.

The contextual menu allows:

• Displaying the comment instead of the name of the nodes
• Displaying the long names of the states
• Copying the tree as an image or as html. The html copy reproduces the vertical display as well as the horizontal one. The displayed colors are also used.

It is possible to perform a manual zoom by clicking and dragging the mouse to select the area that will be zoomed in. The shortcut ctrl + mousewheel performs zooming (and unzooming) on the tree centered on the cursor.

Moving the mouse over a node displays a tooltip containing the chain of evidence done until this node.

A double-click on a node closes its subtrees in the window. Double-click on it to display them again.

The K-S test has been added in the network global performance when the network has learning and test databases.

It is the test of distribution adequacy between the two samples. This test is only computed in the comparison panels. It compares the two distributions of log-likelihoods. The values Z, D and the corresponding p-value are displayed.

In the result panel of the network global performance, a button Extract Database allows extracting the data on which the computed log-likelihood is inside a given interval. The parameters are:

In the Influence Analysis wrt the Target Node a checkbox Fix References becomes available if Inverse Influenceis selected. It allows keeping the value of the a priori probability (red line) as reference if new evidence is set on the monitors. When this option is chosen, a priori is replaced by reference in the caption. This option is also available when data are used. The computed delta values take into account the reference probabilities as well.

The mutual information is computed for each arc of the network (from the Analysis menu or the shortcut J). It takes into account the filtered states.

The thickness of an arc is directly proportional to the mutual information. Three values are displayed in the arcs' comments:

• In black, the mutual information of the arc.
• In blue, the relative mutual information in the direction of the arc.
• In red, the relative mutual information in the opposite direction of the arc.

You can use the slider to change the arc display threshold:

•  Go back to the previous threshold.
•  Go to the next threshold.
•  Store the mutual information in the comments of the arcs if those comments are displayed.
•  Stop analysis.

If all the arcs of a node became transparent, the node becomes transparent.

Now the filtered values are taken into account when mutual information is computed (Target Dynamic Profile, Optimization Tree, Interpretation Tree, Relationship Analysis, Target Analysis, etc.).

In influence analysis, the borders now take the color of the corresponding nodes. Those colors are displayed if the button used to Display Colored Tags of nodes is pressed in the main toolbar.

Now, in all analyzes, the translucent nodes are skipped.

In the target report, filtered states of the dependent nodes are skipped in the analysis.

In the relationship analysis, if some arcs are translucent, because one of their nodes is translucent, they are not analyzed.

This report allows computing the G-test and the independence probability between two variables of the network bounded by a path of length one or two and which is not a V-structure (Analysis menu). All the existing paths of length one or two (without V-structure) will be tested. The independence probability will be computed and only the paths which are not dependent are kept in the report.

Computing independence between variables:

• G-test: The value of the independence test G is computed from the data associated with the network between each pair of variables which are the ends of the paths.
• Degree of freedom: Indicates the degree of freedom between the ends of each path.
• p-value : Represents the independence probability of the G-test between the ends of each path.

The target state optimization has been transformed to become the target optimization. It is now possible to optimize a state of a node or the mean of a node. This can be done in the Target section.

The parameter dialog allows choosing to export the results in an evidence scenario file in the Output section

At the end of the analysis, a dialog indicates that the analysis has ended.

This analysis report is available for any target node (by the Analysis menu). It allows computing the impact on the target state when slightly modifying one propability of a node's CPT.
This is done for each cell of each concerned conditional probability table.
The sensitivity function used is: P(h|e)(t)=P(h,e)(t)/P(e)(t)
where h represents the hypothesis (Target Node = Target State), e represents the context and t the probability.
The derivative of this function is computed as well as the minimal and maximal values when setting the probability to 0% and 100%.
This HTML report has several sections:

1. Context of the analysis
   Description of the observed variables when the analysis is carried out.
2. Probability of the target state
   Probability of the target state knowing the observed variables (context).
3. Maximal Derivative
   This table summarizes the nodes with a maximal derivative different from zero. It is ordered from the greatest to the smallest derivative.
4. Insensitive Nodes
   This table contains the nodes with a derivative equal to zero.
5. Tables of sensitive nodes
   For each sensitive node, a table represents each combination of its values and its parents' values.
   Each cell contains two parts:
   • The part on the left indicates the derivative computed for each combination of states. The background of the maximal derivative is yellow.
   • The part on the right shows the probabilities of the target state when the probability of the a current node's state is set to 0% and to 100%. Obviously, those probabilities are symmetrical for binary nodes.

Monitors

A new menu Monitor has been added to the menu bar. This menu is only visible in validation mode. It contains several items which are the same as in the contextual menu of the monitor panel and in the monitor toolbar.

Now, the log-likelihood corresponding to the current evidence is computed and displayed in the monitor information panel.

Two menus has been added to the menu Monitor and the contextual menu of the monitor panel:

• Sort by Less Probable Evidence: displays the monitors corresponding to the observed nodes ordered from the less probable evidence (that degrades the most the joint probability) to the most probable evidence. Considering that the current network represents mainly the "normal cases", this tool is then very useful to diagnose the variables that have caused a very low joint probability (atypical cases wrt the given network).
• Sort Monitors by Name: displays the monitors under increasing lexicographic order of their names or comments (if comments are displayed instead of names).

Two menus has been added to the menu Monitor and the contextual menu of the monitor panel:

• Fit Monitor Widths to Content: enlarge the monitors in order to display the whole content without cropping the names. This action is done only on the displayed monitors. If a new monitor is displayed and its width is greater, you must click on it again.
• Restore Default Monitor Widths: restore the witdh of the monitors to the default value.

In addition to the option in the settings, the menu Display Probabilities with Scientific Notation is added to the menu Monitor and the contextual menu of the monitor panel. It displays the probabilities in scientific notation in the monitors and in the monitors' tooltips. This option is saved with the network.

Values in the information panel are displayed in the tooltip with a greater number of decimals.

Two menus has been added to the contextual menu of the monitors:

• Fix Probabilities: allows fixing the probabilities of the selected nodes to the current marginal probabilities. Hard observed nodes are not modified but probabilities of soft observed nodes are fixed. This is done on all selected monitors.
• Relax Probabilities: allows unfixing the probabilities of the selected monitors with observed probability distributions. The corresponding nodes become soft observed. This is done on all selected monitors.

The various display properties of the monitors are now saved with the network.

Interface

The task progress bar thus has three states:

•  when the system is ready
•  when a task is in progress. The bar represents an indication of the time spent (note that time evolution is not necessarily linear or even growing). A click on the red light stops the progress of the task.
•  when a task is stopped. The button becomes orange while the process ends. It allows the network to go back to a correct state after having stopped the current task. At this time, the button is no more clickable until it turns green.

Now, in the various html reports, the cells displaying the name of a node have a background color which is the node's color (if any).

A new graph is available: the colored line plot. This graph displays the colored line plot of the two selected variables. The continuous variable is displayed along y-axis. The indices of the values in the database are along x-axis. The color variable gives the color of the curve for each index.

Parameters: Select a continuous variable in the first combo-box and a variable for the color in the second combo-box.

Graph: The black lines represent the limits of the vertical node's intervals. When you move the mouse over the graph, the information in the top panel is updated. It displays the coordinates of the cursor in the graph. The total number of displayed values is also indicated. If you click inside a point (or several if the points overlap), a dialog appears containing the rows of the database corresponding the selected points.

You can perform a zoom on the graph by pressing the left button, drag the mouse and then releasing the button: the selected area will be magnified. To remove the zoom, double-click on the graph, the default view will be restored.

The checkbox Intersection displays only the nodes contained in all the selected classes.

It is now possible to import and export dictionaries of properties in the editors of costs, constants, temporal indices, local structural coefficients and state virtual numbers.

It is now possible to sort the columns of the tables in the editors of costs, constants, temporal indices, classes, local structural coefficients, state virtual numbers, mean variations and forbidden arcs. Simply click on the header of the column you want to sort.

Knowledge Elicitation

The mechanism of knowledge elicitation of BayesiaLab allows a set of experts to provide assessments on the conditional probability tables of nodes. A "facilitator" can mediate to ask relevant questions to experts and get their assessments.

The experts must be declared in the network with the expert editor.

Two ways of capturing the assessments are proposed:

• Each assessment is input directly in BayesiaLab.
• Or each assessment is input via the online assessment tool.

When a node has assessments, the icon is displayed at the bottom left of the node. More the disagreement between the assessments is important for this node, more the background of the icon becomes dark.

A slideshow explains the principle of knowledge elicitation in BayesiaLab. It can be found at the following address:
http://www.bayesia.com/en/products/bayesialab/resources/tutorials/bayesiaLab-knowledge-elicitation-environment.php

Assessments can be entered from the tab Probability Distribution in the node editor. Experts must have been already created. In this case, a button Assessment is displayed and becomes activable when a cell is selected. After pressing the button, the assessment edition is done for the selected line.

When an assessment was made for a line in the conditional probability table, the corresponding cells have a green border.
The icon displayed in the cell indicates the importance of disagreement between the experts for this cell. More the icon is visible, more the disagreement is important.

Move your mouse over a cell displays a tooltip showing:

• The minimum given by an expert for this cell
• The maximum given by an expert for this cell
• The number of assessments for this cell

The assessment editor displays the list of assessments for the selected line of the conditional probability table.

A line in this table defines an assessment. This assessment consists of:

• A probability for each state of the node, the sum of probabilities to be equal to 100. The background of the corresponding cells is gray. These values can be edited by double-clicking or by typing the value if the cell is selected.
• The name of the expert who has made this assessment. The expert is chosen from a combo box containing the name of each expert.
• The confidence the expert gives to his assessment. This value is editable.
• The comment the expert makes about his assessment. A tooltip containing the entire comment is displayed when we move the mouse over the cell.
• The time taken to make this expertise. It is measured in seconds. This value is editable. It is 0 by default when the expertise is filled manually. However, it is automatically filled in when the expertise is online.

The button Add adds an assessment by default at the end of the list. This assessment will then be edited manually. It is possible to have several assessments from the same expert, but, for clarity, it is not recommended.

The button Delete deletes the selected assessments.

When an assessment is selected the expert's image is displayed on the right side:

When you press the button Accept of the dialog box, assessments are automatically normalized if necessary, the sum of probabilities to be equal to 100.
Then, the consensus is computed. He will serve as a probability distribution in the conditional probability table of the node for the corresponding row.
This consensus is computed by averaging assessments for each state weighted by the confidence of each expert. The assessment whose confidence is equal to zero will not be taken into account.

The assessment mechanism needs to define experts associated with the network.

An expert is defined by:

• a name that must be unique
• a credibility which is a real between 0 and 1. When an expert has its credibility equal to 0, his expertise is not taken into account. A fully credible expert has a credibility of 1. This credibility weighs its own expertise relatively to others.
• an image for rapid identification. It can be a photo or an avatar for example.
• a comment which describes the expert (eg competence, domain, etc.)

In the editor expert, these four properties can be edited by double-clicking on the corresponding cell. The fifth property displayed is the number of assessments the expert has made in the network. This property is not editable.

The table can be sorted by clicking on the column headers.

The button Add opens a new window allowing the user to enter information about a new expert:

To add or change the image, simply click in the box. The verification of the uniqueness of the name is made during the validation.

The button Remove the selected experts of the list. The assessment done by this expert will also be removed from the nodes. However, the conditional probability tables won't be regenerated.

The button Import allows loading a dictionary of experts. The format of the dictionary is the following:

The button Export exports a dictionary of experts. The images of the experts will be saved in png format in the same directory as the dictionary with the name expertImageN.png

The button Open Session allows opening an online assessment session.

The user msut provide a session identifier (unique) and a password.
Once the session opened, the experts can connect with their browser to the folowing secured address:
https://www.bayesialab.com
The user must enter his name as given in the expert editor (respecting case) and the session identifier.
To create an online assessment session, you must contact Bayesia directly: info@bayesia.com.

If a session with the same name has already been opened, a dialog box will offer to overwrite if this session comes from the same machine. If it comes from another machine, it will not be possible to use this session name.

the button Close Session allows ending the current online assessment session.
Experts which are still connected will be disconnected.
Closing a network or BayesiaLab while a session is running automatically closes the session.

This button is active if at least one expert is selected. It generates conditional probability tables of the nodes by taking into account the assessment of selected experts. If none of the selected experts has made assessment for a node then the consensus of all experts is used. The credibility of experts is used for generation

When at least one expert is associated with the network, the indicator   is displayed in the status bar of the network. A clic on this icon displays this editor.

When an online knowledged elicitation session has been created, a button Post Assessment is displayed in the window.

When you press this button, the current question is posted online so that experts can answer it connected via a dedicated web interface. This question asks the experts to get the probability distribution that corresponds to the chosen line of the conditional probability table.
If assessments already exist for this question, the web interface for relevant experts will be pre-filled.
A window displays, in real-time, the experts who anwsered the question:

This table can be sorted by clicking on the header of each column.

Once the online session has been opened, an expert can connect with its Internet browser to following secured address:
https://www.bayesialab.com

He will enter his expert name and the name of the session respecting upper case:

Once connected, the waiting interface will be displayed:

Once a question is posted through BayesiaLab, it will appear in the interface of each expert. If the expert has already answered this question, the corresponding fields are already filled. They may be modified if necessary.

On the right side, the name and the comment of the node are displayed. A colored slider and a field allows editing the probability associated with each state the node.
When the probability of a state is changed, the others will be adjusted in order to always sum to 100.
A lock located on the left allows blocking the corresponding probability so it is no longer editable. The probability can be unlocked by clicking again on the lock.

The expert can also enter the confidence it grants to its assessment.

A text field lets you enter the comment associated with this assessment.

On the left, the context of the question is displayed, i.e. the current states of each parent of the node. By hovering the name of the parents, their comments will be displayed in a tooltip.

The pie chart in the bottom left reproduces the current probability distribution.
A textual indicator indicates in a simplified way the confidence of the assessment.

The button Validation sends the anwser of the expert to the server then the indicator of experts who have answered is updated on BayesiaLab. After that, the waiting interface is displayed again.

The contextual menu of the expert indicator allows us to generate an assessment report.

The menu Tools > Assessment provides access to various tools for exporting assessments.

The menu Analysis > Graphic provides access to the assessment sensitivity analysis.

This tool allows, through sampling, visualizing graphically the impact of the different assessments of network's variables on each state of target variables. This tool is close to the parameter sensitivity analysis. The following dialog box lets you choose the parameter nodes and target nodes and to indicate the number of samples to be used:

The icon indicates that the node has assessments.

Three policies are available for sampling:

• Random selection of one expert per network: for each sample an expert will be drawn at random and the conditional probability tables will be generated only with the assessments of the expert.
• Random selection of one expert per node: for each sample and for each node an expert will be drawn at random and the conditional probability tables will be generated accordingly.
• Random selection of an assessment per parent combination of each node: that is to say randomize an assessment by row of the conditional probability table. For each sample and each row of the table of each node, an assessment is used to generate the probabilities of this line.

In each case, if there is no expertise that matches the sample, then the consensus is used.

For each target node, the result of the analysis is presented with a curve representing the repartition function of the probabilities of each state, and, a bar chart representing the probability density function. Besides these graphical results, the mean and the standard-deviation of the probabilities of the target states are also given.

The user selects the node and the state to see by clicking on their tabs.

In this example, we have 32% chance that the state "Weak" of the node "TARGET" has a probability between 70% and 72.5%.

The analysis is performed over the whole nodes or over a subset of selected nodes. The translucent nodes are not taken into account. The context of the evidences is taken into account and displayed under the graphs.

A contextual menu allows displaying the comment associated with the nodes instead of the name and also the long names of the states. It allows copying the chart as an image.

This report is generated from the contextual menu of the expert icon in the status bar.

It contains two parts. The first part is a global summary analysis of the assessment:

It shows the list of experts: their name, credibility, comment and the number of assessments done like in the expert editor. The last column shows the average assessment time used by each expert to make assessment. This time is computed only when an online session is used.

After that, two tables are displayed:

• The first table is the list of the nodes with assessment. The first column is node name, the next column is the comment of the node (depending on the settings) and the last one is the global disagreement of the assessments. This percentage represents the average deviation of each assessment with respect to the mean of each cell. It takes into account the confidence associated with each assessment. If an assessment has a confidence equal to zero, it won't be taken into account in the global disagreement.
  The nodes are sorted according to the global disagreement.
• The second table is also the list of the nodes with assessment. The first column is node name, the next column is the comment of the node (depending on the settings) and the last one is the maximum disagreement of the assessments. This percentage represents the maximum deviation of all assessments in the whole table. If an assessment has a confidence equal to zero, it won't be taken into account in the maximum disagreement.
  The nodes are sorted according to the maximum disagreement.

The global and maximum disagreements between the experts allow us to easily find on which nodes the knowledge of the experts is not the same and the knowledge elicitation should be verified.

The second part of the report details the assessment for each variable:

For each node with assessment, a table contains:

• The number of rows in the conditional probability table that have assessments associated with compared to the total number of rows
• The total number of assessments done on this node
• The number of experts involved compared to the total number of experts
• The global disagreement of the node
• The maximum disagreement in the conditional probability table
• The global assessment time which is the sum of all assessment times
• The mean assessment time by row of the conditional probability table
• The mean expert assessment time for this node

Tools

A submenu Assessment is available only if experts are registered on this network and assessments have been done. It gives access to:

• Export a Network per Expert: the user select a destination directory. For each expert, a network with the same structure but with the probabilities given only by this expert will be saved in this directory. When there is no probability given by this expert, the consensus of the other experts is used.
• Export Probability Assessments: exports into a database all the assessments done by the experts. The database contains a column for each node with assessments, a column with the correponding probability given by the expert, a column for the confidence of this assessment and also the name of the expert and the time used to do this assessment (see Assessments).
• Export Expert Assessments: exports into a database for each cell of each table with assessments, the probability given by each expert and the associated confidence. Then, there are two columns by expert, one for the probability and the other for the confidence. A column Weight give a weight to each row corresponding to 1 / number of states (see Assessments).

In cross-validation tools, data are shuffled before slicing the database in order to avoid biases due to ordered data.

This tool is used to compare the joint probability distributions from two different files. These files have been previously generated by the Batch Joint Probability keeping only the computed joint probability in the destination file.

This analysis is similar to the Global Network Performance but on the joint probabilities present in the input files.

The files must be indicated in the following panel:

For each panel, the mean, standard deviation, minimum, maximum and computed row numbers are indicated below the chart.

In a first time, each file of joint probabilities is analyzed separately and after that the comparison is done.

The Kolmogorov-Smirnov Test is computed to mesure the distribution adequacy between the two samples. this test is present in the comparison panels only. It compares the two distributions of log-likelihoods. The values Z, D and the corresponding p-value are displayed.

In cross-validation tools, the structural coefficient of the reference network is recomputed for the generated networks by multiplying it with final learning weight divided by the inital learning weight.

In cross-validation tools, if the initial databse is stratified, the generated databases will be also stratified with the same parameters.

The menu item Tools>Cross validation>Structural Coefficient Analysis allows computing a frequency value that measures how robust is each arc. The network needs a database and the validation mode must be activated.

 Warning : Use only when data is scarce when used with low structural coefficient.

For analysing the structural coeficient, the chosen learning algorithm will be tested with different values of the structural coefficient in a given interval. At each iteration, the network structure is learned on entire database with a growing structural coefficient.

Simply select the learning algorithm we want to use, and indicate the minimum and maximum limits for the structural coefficient and the number of iterations to perform in the following dialog box:

The structural coefficient can vary from 0 to 150.

Three options are available allowing computing at each iteration:

• Structure/Data Ratio: use it for unsupervised tasks. It is the ratio between the natural structural complexity (with a coefficient to 1) of obtained networks and their data likelihood (with a coefficient to 0).
• Target's Precision in %: use it for supervised tasks. A target node is necessary. The precision of the target prediction is computed for each obtained network.
• Structure/Target's Precision: use it for supervised taskstaking into account the structure complexity and the precision. A target node is necessary.

The tested structural coefficients will vary from the given minimum to the maximum by step of (maximum - minimum) / number of iterations.

An output directory can be specified where all networks learnt will be saved.

Depending on the chosen learning algorithm, a dialog box displays specific settings:

Once the networks have been learnt on each structural coefficient, the following report is displayed:

This report is similar to the Arc Confidence except for the last table where the column indicating the maximum structural coefficient of the structure has been added. This indicates, among the coefficients tested, which is the greatest that has obtained this structure.

The report can be saved in a HTML format file. It can also be printed. Three other options exist: displaying graphs, extracting the network and displaying the curves.

The Graphs button from the report allows displaying the graphical structure comparator. With this tool, data contained in reports can be viewed and interpreted easily.

The Network extraction button from the report displays network extraction tool. This tool allows building a network from any structure depending on arcs frequency thresholds.

The button Curves of the report allows displaying the curves corresponding to the options chosen in the parameters. A dialog box allows the user to choose the curve to display:

• Structure/Data Ratio : the structural coefficient can be chosen in the area at the inflection point of the curve, just before a strong increase by reading the graph from right to left.

  

• Target's Precision in % :

  

• Structure/Target's Precision Ratio :

  

The menu item Tools>Cross Validation>Data Perturbation allows avoiding the local minima of the network's structure in which the network can be trapped. The network needs a database and the validation mode must be activated.

The BayesiaLab's structural learning algorithms are based on heuristic search. Then, they can be trapped into local minima. As they are based on different heuristics, those local minima can be different. Applying every algorithms and choosing the one with the lowest score represents the first solution to optimize the obtained network. This data perturbation tool provides another solution consisting in adding noise to the weights associated to each line of the database in order to try escaping from local minima.

A noise is generated by using a Gaussian law, with 0 mean and the standard deviation set by the user. The selected learning algorithm is applied on this perturbed database and the score of the final structure is computed by using the original weights. The decay factor is applied after each iteration to reduce the standard deviation.

Simply select the learning algorithm we want to use, and indicate the initial standard deviation, the decay factor and the number of tests to perform in the following dialog box:

An output directory can be specified where all networks learnt will be saved.

Depending on the chosen learning algorithm, a dialog box displays specific settings:

Once the networks have been learnt, the following report is displayed:

This report is similar to the Arc Confidence except for the last table where the column indicating the Structure's Mean Score has been added.

The report can be saved in a HTML format file. It can also be printed. Two other options exist: displaying graphs and extracting the network.

The Graphs button from the report allows displaying the graphical structure comparator. With this tool, data contained in reports can be viewed and interpreted easily.

The Network extraction button from the report displays network extraction tool. This tool allows building a network from any structure depending on arcs frequency thresholds.

This function can be found in the Tools menu. The network must have a database and a target node. This tool allows analysing a network having a variable called "selector" which must be different from the target node. for each state of this variable, one of the three available analysis will be performed on the variables of the network with respect to the target.
In this way, it is possible to analyze the behavior of each variable with respect to the target for each state of the selector variable and to compare those behaviors between them.

The classic use of this tool is when the variable "selector" is a list of products we want to compare with each other to analyze their strengths and weaknesses relatively to the target variable.
In the example we use, the variable selector (Product) represents a range of perfumes and the target variable is the Purchase Intent of these products.
Other variables that will be analyzed in relation to the target represent the qualities of these perfumes.

The translucent nodes are not taken into account in these analysis.

For each state of the selector, the algorithm will create a network with the same structure as the original network and with a database containing only the lines corresponding to the current state.
The probability tables are learned with these new data. Then the selected analysis is performed on each network.

The following panel lets you define the parameters of the analysis.

• Selector Node: you can choose from a drop down list the variable that will be used as selector node. The target node is excluded from this list.
• Analysis: allows choosing the analysis among:
  • Mutual Information: compute the conditional mutual information between each variable and the target
  • Total Effects: compute the total effects of the variables on the target
  • Standardized Total Effects: compute the standardized total effects of the variables on the target
• Linearize Nodes' Values: by checking this option, the values associated with states of nodes are recomputed and sorted in order to have a positive increasing impact on the value of the target variable.
• Regenerate Values: by checking this option, the values associated with states of continuous nodes are re-computed from the data in each database for each state of the selector variable. If the linearization is requested, it will be done after the generation of values.
• Output Directory: allows specifying a directory in which each network corresponding to each state of the selector will be saved.

The display of the results takes the form of points positioned on a 2D chart. It is similar to quadrant chart in the correlation with the target node report or in the total effects report.

The points represent the variables. If a node has a color then the point will be displayed in this color. The name of the node is displayed at the right of the point.
When moving the mouse over a point, the coordinates are displayed in the top panel.
The top panel shows the number of points displayed as well as the evidence context.
It is possible to zoom in the chart as in the scatter of plots.

The mean value of the nodes is displayed along x-axis and the reult of the chosen analysis is along the y-axis.
The title of the chart indicates the performed analysios and the current state of the selector node. the states can be selected via the chart's contextual menu.

The chart's contextual menu allows to :

• Display the comment of the nodes instead of their name
• Display the long names of the states
• Display the scales
• Select the state to display by choosing one in the proposed list
• Copy the chart as image or array of points (text or html)
• Print the chart

The chart above represents the value of each node according to the chosen state. When we move the mouse over a node, other nodes disappear and the values of this node for each state of the selector are displayed. In the example, the relevant node is "Tasty" for the state 8 of the selector node "Product". Other values of "Tasty" are displayed with the name of the corresponding state next to them.

When choosing the option Display Scales from the contextual menu, chart displays for each node a scale between the minimum and maximum values among all states of the selector. The mean is also indicated by a vertical line. It is therefore possible to easily identify where is the value of the node for the current state among all other states. So we can identify the possible negative and positive progress margins for the current product for each of its qualities.

By hovering the mouse over the point, we obstain, as previously, the different values of the node for each state in addition to the scale:

The button Export Variations allows exporting the percentages of negative and positive variations of each node in order to use them in other analysis thanks to the mean variation editor.

Settings

• Maximum Cluster Size: indicate the maximum number of clusters automatically chosen by the algorithm.
• Stop Threshold: indicates the threshold corresponding to the maximum KL weight a cluster can have to be kept.

This item is visible only if your license allows it. It is also not available for Mac OS X.

When an instance of BayesiaLab is already running, if the user wants to open again BayesiaLab or performs a double-click on a XBL file, the previous instance will be automatically reused. The following options allows managing this behavior:

• Allows opening multiple instances of BayesiaLab: when an instance of BayesiaLab is already running, it is possible to open a new instance if the license allows it. Opening a new instance can be done either by running directly BayesiaLab or by double-clicking on an XBL files for example (depending on the second option). It works as well with command line on any platform. If the option is unchecked, opened instance will be always reused and the second option will be always checked.
• Reuse opened BayesiaLab when double-clicking on XBL files: when an instance of BayesiaLab is already running, it is possible to reuse it when the user performs a double-click on an XBL file (or by command line). If the first option is checked and this option also, the user will open a new instance of BayesiaLab if he runs directly the program and he will reuse an already opened instance if he double-clicks on a XBL file. It works as well with command line on any platform.

BayesiaLab needs to be rebooted to take changes into account.

• Use by Default Scientific Notation for Probabilities: when checked probabilities are displayed by default with scientific notation in monitors and with percentage in the corresponding tooltips.

Perspectives allows the user to configure the different menus of BayesiaLab as he wants. Each menu of the menu bar is present in the column Component and can be set visible or not if the user doesn't need it. This can be done by checking of not the corresponding option in the column named Visible.
It is also possible to configure the keyboad shortcut for each menu item. The column Shortcut shows the current shortcut associated with the menu item.
Some of components cannot have their visibilities modified, like Settings and Perspectives for obvious reasons. The last column Count shows how many times each action has been done. It is useful to understand how BayesiaLab is used. A special tool allows analyzing these results.

It is possible to edit the shortcut of a menu by selecting the component in the list, for example Target's Optimization Tree:

Once selected, the buttons and field in the bottom part become available.
After that, you can choose none, one or several modifiers among Ctrl, Meta for Mac users, Alt, Shift and Alt Graph.
Now click on the field and type the key you want for the shortcut. In our example, we want to create the shortcut Ctrl + O. So the button Ctrl is pressed and the key O is typed in the text field.
The interface indicates that this shortcut is already used for the Open menu, so we must change the keystroke by pressing the button Alt in order to obtain Ctrl + Alt + O.

To remove a shortcut, select it and click in the text field. Once the focus is on the text field press the key or (not the buttons).

The button Load allows importing a perspective file. The button Save allows exporting the current perspective into a text file.

To validate all the changes, you must click on the Apply or Accept buttons.

Help

This analysis can be accessed by the menu Help > Use Analysis.

With the registration of the use of each menu of the BayesiaLab's menu bar (see Perspectives), it is possible to make graphical analysis of the use of BayesiaLab's tools.

The window contains in its right hand side the tree of BayesiaLab's menus. Each item can be selected and the chart of use appears in the right side as a pie chart. Pie charts can be displayed under two forms that can be selected by the options on the right:

• Use percentage indicates the number of used functions and the number of never used functions.
• Detailed use indicates the number of times each function was used.

The following figure shows the overall use of BayesiaLab, i.e. the number of functions used (red) and the number of functions never used (green). Just move the mouse pointer to display a tooltip information corresponding to the specified area.

When you select a menu or submenu in the tree on the left, the graph shows, then, only the use of the selected menu and its submenus. In the following figure the Analysis menu is selected:

To see the detailed chart, simply press the button Display Detailed Use which displays a window containing a larger as well as the detail of the use on the right hand side.

When you click on the second option, the chart displays the number of times each function was used, according to the menu selected in the tree:

To see the detailed chart, simply press the button Display Detailed Use. The right part shows for each sector the number and percentage of use compared to the sum of all uses.

The button Display Table opens a new window containing the table all BayesiaLab's menus with their name, their icon, their visibility, their shortcut, and their number of uses. When a row is selected, pressing on the button help displays the help corresponding to the function, if it exists.

the table is sortable by clicking on each column's header.

Shortcuts

In all modes, a double-click on a node while pressing T sets the node as target with its first state as target state.

In validation mode, a double-click on a probability bar of a monitor while pressing T sets the corresponding node as target with the corrresponding state as tarrget state.

An node can be set as excluded or not excluded by double-clicking on it while pressing X in modeling mode only.

The default shortcut for total effects is Shift + T.

When the focus is in the graph area (by clicking on a the background, for example), you can zoom in and out with Ctrl + Mouse Wheel.

When the focus is in the monitor area (by clicking on a monitor, for example), you can zoom in and out with Ctrl + Mouse Wheel.

Ergonomics

Now BayesiaLab can be reused if it is already opened by running the program directly or by double-clicking on an XBL file.

If the program is uninstalled and is currently running, BayesiaLab will be automatically closed.

The Mac OS X standard application menu has been added. Moreover, BayesiaLab can be opened by double-clicking on XBL files and can be reused if it already opened. BayesiaLab now acts as other applications on Mac OS X.

The setup file without JRE can be installed on any Windows 64 bits. Installing a 64 bits JRE before is required.

The bug report mechanism now uses the default user's mailer program to send the error log. The user can modify the generated e-mail if wanted.

Miscellanous

In addition to the standard, professional and educational editions, a new edition named analyst is now available. The analyst edition is only oriented towards network analysis. It does not contains data-mining tools.