FR2995421A1 - Method for performing navigation in navigation tree in database, involves associating nodes with level, and executing set of navigation controls to select one node in tree and display information associated with selected node - Google Patents

Abstract

The method involves selecting a set of classification elements, and attributes associated with the set of classification elements. A navigation tree is calculated from a set of objects selected by a user and from a strategy from a menu zone (Z7) so as to order a set of levels of the navigation tree. A set of nodes associated with a level and a value of the attributes is displayed on a screen of a terminal. A set of navigation controls is executed to select one node in the tree and to display information associated with the selected node.

Description

A navigation method within a navigation tree created from data relating to elements. FIELD OF THE INVENTION The invention relates to the calculation of trees and navigation within a calculated tree and a set of calculated trees. 2. PRIOR ART In the field of databases, an important issue is to visualize important elements for a user while processing a considerable mass of information. One technique for reducing the amount of information to display lies in the tree structuring of the data. The lowest level of the database contains the basic element, for example an individual, a company, a document, a city, and generally any entity that can be evaluated according to several variables. These basic elements are associated with attributes according to a variable which constitutes a criterion of evaluation of the element. An attribute is a value of a variable applied to a basic element, for example, the attribute of an individual according to the variable "age" is its number of years, a book is associated with the attribute "adventure" according to the variable "gender". The database can be represented by a table with each row associated with a base element, and each column an attribute. The information of such a table is not exploitable visually especially if there is a large number of elements and evaluation variables. One way to show the information is to order the variables, to display a tree in which each level represents a variable, and to present as much departure from a node as there are attribute values associated with that node. A user has a screen and keyboard to navigate the tree representation. The menu includes an area showing the structure of the tree with an indication of a current node, and a field listing all the basic elements accessible from the current node. By navigating from one level to a lower level, the user selects an attribute and thereby restricts the number of accessible elements. At the top of the tree, all the basic elements are accessible, and the more the user goes down to lower levels, the more the number of accessible elements decreases. We can consider that the last level consists of the basic elements. This representation obliges the user to define himself the parameters of the tree and in particular the scheduling of the variables defining the hierarchy of the nodes. As a result, the user must first know the importance of certain variables to schedule them, which is not always the case. There is therefore a real need for a tool allowing a user to discover data analysis axes in order to determine the relevant variables for analyzing a situation. In addition, the tree representation software using all the information stored in the database, the used memory size becomes considerable, and tree modeling can limit the size problem. Finally, the creation of the navigation tree according to the parameters defined by the user requires calculation power proportional to the amount of information to be processed. In the case of a database containing a large number of variables, a user does not always have a large computing power. There is therefore a real need for a tree creation module based on information extracted from a database, making it possible to navigate within a representation. 3. Objectives of the invention The present invention provides a solution that does not have the drawbacks described above, while providing the advantages listed above. 4. DESCRIPTION OF THE INVENTION Functionally, the invention relates to a navigation method within a navigation tree created from data relating to elements, these data being associated with attributes according to classification elements. or variable. The method comprises the following steps: - selection by an operator of classification elements and attributes associated with classification elements, - calculation of a first navigation tree from the objects selected by the user and from a navigation strategy for ordering the levels of the navigation tree, - displaying on a screen of a terminal of said first tree by displaying nodes associated with a level and an attribute value, the links between the nodes displaying the ordered levels of the first navigation tree, - executing navigation commands for selecting a node in the first tree and displaying information associated with the selected node. In this way, the user can navigate a displayed tree and refine the display by selecting some nodes and browsing the levels. According to a first improvement, the method compares a value of a given indicator from at least one attribute and at least one threshold, and displays a graphical distinction for each node of the first tree that varies according to the difference between that value of a given attribute and at least that threshold. In this way, the user can refine his navigation by focusing on a particular attribute. According to another improvement, the value of the indicator is determined from a combination of attributes. In this way, the user can evaluate the relevance of a combination of attributes on the set of attributes and know which attributes have the most influence on an element. According to another improvement, the method comprises a rotation control of the representation of the first navigation tree. In this way the user can display on the screen of his terminal a part of the tree and by rotations take cognizance of all the nodes and links connecting the nodes. According to another improvement, the user can select a node defining a set of elements accessible from this node. A new navigation tree is calculated using the navigation strategy applied to the set of elements thus defined, and displayed. In this way, the user can work in a subset of easily selectable items. According to another improvement, the first tree and the new tree are displayed simultaneously, and the user enters a command to navigate alternately from one tree to the other displayed tree. In this way, the user easily switches from the set of elements to the subset. According to another improvement, the calculation of the first navigation tree is performed at a remote server, the terminal then transmits the creation data of the tree to said server. In this way, the database and the processing for the calculation of the trees can be deported. 5. List of Figures Other features and advantages of the invention will appear more clearly on reading the following description of an embodiment. particular, given as a simple illustrative and non-limiting example, and the accompanying drawings, among which: - Figure 1 shows an example of a screen appearance of a configuration menu for the creation of a navigation tree - Figure 2 shows an example of a menu for the determination of an indicator and the launch of at least one tree based on this indicator, - Figure 3 shows an example of navigation menu with a first tree and three levels, FIG. 4 shows an example of a navigation menu with a first tree and 4 navigation levels; FIG. 5 shows an example of a navigation menu showing a first and a second navigation tree; avigation. - Figure 6 shows an example of a navigation menu showing the relevance of an indicator for each node. 6. DESCRIPTION OF AN EMBODIMENT OF THE INVENTION 6.1 GENERAL PRINCIPLE The present invention relates to the creation of navigation trees from data relating to elements, these data being associated with attributes according to classification elements. or variable. The navigation tree thus calculated comprises a plurality of levels each associated with a classification element, a plurality of nodes being present in a level, each node being associated with a given attribute. An initialization step allows an operator to define the classification elements and attributes associated with each node. The operator can define an indicator whose value is compared for each node to evaluate its relevance. The invention also relates to the display of a tree structure by highlighting nodes and links uniting these nodes and, navigation on a navigation structure, navigation commands are introduced to select nodes appearing on the tree. navigation tree, to display a new tree level or to initiate the opening of a new subtree from the current node of a primary main tree. Associated graphics distinctions are displayed when node values are below or above a certain threshold. 6.2 Description of an Embodiment Principle The tree creation module which is an object of the present invention uses the information stored in a database. This database contains base element identifiers associated with attributes based on a plurality of variables that constitute element evaluation criteria. A basic element can be an individual, a document, a company, and generally any entity that can be evaluated according to several variables. An attribute is a value of a variable applied to a base element. For example, in the case of a structure whose basic elements are individuals, it is possible to evaluate each individual by the variables and the list of attributes associated with each variable: - sex (attributes Man, Woman) - age (1-15 years old, 16-30 years old, 31-45 years old, 46-60 years old, 60+ years old) - monthly income (- 500 E, 500 to 999 E, 1000 to 1999 E, 2000 to 2999E, 3000E and + - number of dependent children (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, more than 10) - activity (in training, job search, active, retired) - number of visits per year to a physician (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, more than 10) For multimedia documents, other variables and attributes exist such as: - Language: English, French, German, Italian, Arabic, others - Theme: film, information, documentary, sport, game / show Second level of the theme (associated sub-attribute): the sub-theme : - movie: {adventure, drama, erotic, detective, fiction, others} - inform ation: {newspaper, radio, weather, stock exchange, sport, other} - documentary: {general, nature, culture, social, history} - game / show: {general, variety, circus, talk show, circus, theater} - sport : {General, football, tennis, basketball, athletics, nautical} - Source: TPS, Canal Satellite, DVD lounge, VCR, Web. - Morality: 0 (all public), 1, ...., 3 (pornographic). - Media: audio, video, image, text, - age of the document, measured for example not its date of creation.

The information associated with each base element is stored in a database. Each basic element is identified by a string or a specific number, followed by a list of attributes. All the information is accessible by at least one server responsible for establishing analyzes and helping a user in his decision-making A user requests an analysis using a parameter sheet created during a configuration step . This form can be a paper sheet sent to the database manager, or an electronic form that is displayed on the terminal of a user. By calling the form, a configuration menu appears on the screen. This menu can take many different appearances, and have a greater or lesser number of windows. Figure 1 shows an example of a screen appearance of a configuration menu. In this example, the user has a mouse to move an index on the screen, and function keys to select an icon or to make it disappear. In the example illustrated in FIG. 1, the menu comprises a zone Z1 listing the available variables. According to the example, the focus is on the variable age, referenced 13. A second zone Z2 displays the details of this variable, for example global information such as the average age of all individuals including the attributes associated with this variable . Zone Z2 also displays the list of attributes associated with the variable placed in the focus, in this case all age groups. Additional information is displayed for each attribute, for example the number of individuals for each age group. A third zone Z3 presents a summary of the parameters already introduced by the user. According to the exemplary embodiment illustrated, in this third zone, it can be seen that the user has already programmed 5 levels. Delete, insert and change level commands are provided and accessible from the interface. In this menu, the user has the possibility to select certain attributes to be present in his navigation tree. In the example shown in Figure 1, the user has asked to consider only three age groups for the age variable, and the product navigation tree will display only 3 branches from the "age" node. . A fourth zone Z4 presents general information about the current configuration, for example the number of levels already introduced by the user. This fourth zone has an icon for determining an indicator to discriminate. By clicking on this icon, the user accesses an indicator determination menu, an example of which screen appearance is shown in Figure 2. Figure 2 shows an example of an indicator initialization menu. 10 An indicator is used when the user wants to explore a variable and looks for the elements to explain its evolution in order to discover the risks and opportunities associated with this indicator. When displaying the navigation menu, the value of the indicator is determined for each node. This determination can be made from an attribute or a combination of several attributes. The value of the indicator is compared to a threshold defined by the initialization menu. If this value is greater than the threshold, then the node is graphically highlighted. In this way, the user immediately notices the most relevant nodes for the variable he wants to analyze. Note that a graph can display several analysis axes and a single indicator can be evaluated for a given graph. If the user wishes to change the indicator, he must define it and start a new tree calculation. A zone Z5 lists all the variables present in the database, these variables are identified by a number and a string illustrated by a letter in FIG. 2. A zone Z6 allows the user to select the variables it wants to analyze in order to know their influences on the indicator being initialized. From the selected variables, the tree creation module generates as many navigation trees allowing the user to visualize the influence of each variable for this indicator. The purpose is to discriminate the values of an indicator based on the variables selected by the user.

The Z7 area of the menu in Figure 2 allows the user to select one or more tree creation strategies, by ticking a box. A first strategy is to randomly order the variables. A second strategy consists in ordering the variables according to the number of attributes of each variable. Another strategy is to choose by ordering the variables whose attribute has the highest concentration of an indicator. Depending on the selected strategy, it can be seen that the tree generation module analyzes the data associated with each base element to select variables and attributes and assign them to levels and nodes. Whichever policy is selected, the first level of the navigation tree starts from the variable selected by the user and corresponding to that tree. For example, the user defines an indicator that represents the dangerousness of diseases towards individuals, the wording of this indicator appears at the top of the menu of figure 2. Suppose that the analysis is done in a database of which the basic elements are individuals. The user selects a number of variables to analyze against this indicator. The tree creation module will highlight the fact that the value of this indicator is directly proportional to the age (variable 13) and the number of visits per year to a doctor (variable 7). As a result, the user can analyze the "Disease Risk" indicator for each individual based on these two variables. With the help of the graph analysis tool, it is possible to refine the calculation of the "Disease risk" indicator by taking into account the "visit number" and "age" attribute values. In the Z8 area of the menu of Figure 2, the user introduces the graphic distinction showing value thresholds of the indicator. For example, for values of the indicator lower than or equal to a first threshold ("50" according to the example of Figure 2), the node has no graphical distinction. For values between this first threshold and a second threshold ("50" and "80" according to the example of Figure 2), the node is displayed in orange.

For values of the indicator greater than the second threshold ("80" according to the example of Figure 2), the node is displayed in red. Other graphic distinctions are possible in the context of the present invention, such as: blinking of the node icon more or less fast, size of the icon of the node more or less large, brilliance of the icon more or less Important, Underline, In the menu illustrated by Figure 2, the user can select one or more strategies from those proposed. Without user selection, this is a default policy that is used by the navigation tree creation module. Principle of calculation of trees Once the user has finished initializing the parameters defining the indicator of the graph by listing the variables to be taken into account to analyze this indicator, he selects the icon "Output with calculation of trees" ". Selecting this icon starts the operation of the input data and the creation of the navigation trees. The tree creation module generates as many trees as variables selected in zone Z6. In addition, if the user has selected multiple strategies for the same variable, the module will generate as many trees. By viewing the different trees and navigating on them, the user determines the discriminatory variables next to the indicator selected by the menu of Figure 2. The tree creation module uses the data entered in the initialization menu of the tree. the indicator for creating a plurality of navigation trees. Navigation data occupies a much smaller memory space than that of the database itself. According to a preferred embodiment, the terminal of the user transmits the indicator initialization data to a remote server, which responds by transmitting the data of the plurality of navigation trees. In this way, the user does not need to have all the information in the database, nor the computing power needed to create the navigation trees. Once downloaded to the user's computer, the user launches the navigation module that retrieves the navigation tree data, decompresses the data and displays a representation of a tree calculated by the server. Navigation and initialization commands Figure 3 shows an example of a navigation menu displayed on the user's screen. In this example, the screen is divided into two areas, one on the left shows the navigation tree on a limited number of levels, three in this case, and on the right is an area displaying all of the basic elements. accessible from the current node (the basic elements are identified by Greek letters in Figure 3). When the menu appears, the current node is the first level node. In Figure 3, the identifier of the node of the first level is "A" and its associated parameters are: 125, 5 and 10 ° A. The selection of the parameters to be displayed for each node in the navigation menu is performed by use in a submenu of the configuration menu. The current node is identified by a graphic distinction, such as a different color, a specific size, or a blink. Figure 3 shows a current node at the second level of the tree, its identifier is "E" and the parameters associated with this node are: 89, 13 and 30 ° A. The navigation commands are accessible from a user interface such as a keyboard with direction keys and selection and validation keys, or a graphical cursor moving on a screen by the movements of a mouse. computer or hand movements captured by a camera, or sound control analyzed by a voice recognition engine, or any other means of introduction of commands. The user moves his index to position it on a node to make available the commands to be applied on this node. The positioning of the cursor on the node triggers the display of a window presenting the associated general data, for example the identifier of this node. Attribute values associated with a node appear in a window near that node. An action on a mouse key (typically a left click) validates the choice of the node under the cursor that becomes the current node. The list of basic items displayed on the right side of the menu is immediately updated. Figure 4 illustrates this command, the current node is now at the third level, and the list of basic elements shown to the right is more restricted. To better show the interactions between nodes, the user has tree rotation commands. This command is typically accessible by holding down a mouse button, while moving it. The set of representations of the nodes and branches of the displayed tree seems to be shot in 3D, the nodes appearing as posed on the surface of a virtual sphere. The user has a command to create a new tree, from a current node of a first tree, this node corresponding to a certain variable. The command is for example the close support on a key of the mouse. The navigation module then displays a second tree whose original node is associated with the variable corresponding to the current node of the first tree. This second tree then highlights the variable selected by the user, so that the user can easily analyze visually the influence of this variable taken alone on the indicator. Figure 5 shows an example of a navigation menu showing smaller a first tree, and larger a second tree in which is the current node. The first tree remains in the menu and can be selected again by a user command. The first tree retains all the navigation commands performed by the user and in particular the selection of the different nodes. The newly displayed second tree shows three levels, the first of which contains the current node. In this way, the user can navigate from one tree to another and compare the variables of each tree and the data associated with each node of each tree. Tree view menu Levels associated with indicators appear during navigation.

Indicator values are compared to thresholds and graphically highlighted when menus are displayed. Figure 6 shows an example of a visualization menu with an indicator taken into account. Previously, the user was careful to initialize an indicator using a menu whose example is given in Figure 2. The graphical distinctions applied to each node according to the value of the indicator are defined in the zone Z8 of the menu of FIG. 2. The icons represented in FIG. 6 are marked by a white, gray or black circle according to the value of the indicator applied to each node. For example, the indicator is the product of the first two attributes and the thresholds defined by the initialization menu are 1200 and 2000 for example. The value of the indicator of node A is 625, that of node B: 1040, that of node C: 1520, that of node D: 1602 and that of node E: 2352. The nodes A and B do not have graphical distinction, the icons of nodes C and D present a gray circle and the icon of the node E is represented by a black circle in figure 6. These graphic distinctions are particularly useful when the large number of nodes or attributes does not allow not displaying the attribute values of each node. The graphic distinction associated with the indicator value may appear by a specific color, or a highlight. For example, the indicator may express a risk associated with an individual. The variable "risk" which depends on the variable "age" and the variable expressing the number of visits per year to a doctor. If this variable exceeds a predetermined threshold, the corresponding node appears in red on the menu, or the node icon flashes. According to one improvement, the navigation module has a sub-population exploration command called "Drill-Down" in the prototype. This command creates a new set of trees by applying it to a subset of basic elements. For this, the user navigates in a first tree by selecting nodes associated with a certain value of attributes. Basic elements that are accessible from this level, and appear in the box on the right, are a subset contained in the set of basic elements. The basic elements of this subpopulation have among their data the attributes of the nodes crossed by the navigation in the first tree. By activating a command, the user triggers the generation of a new tree whose original node has the same variable as the current node in the first tree by limiting the analysis to the data associated with the subpopulation. The navigation module searches the data of this new tree and eliminates from the display the elements that do not belong to the subset thus determined. The fact of limiting the analysis to a sub-population makes it possible in particular to reduce the number of accessible attributes and to focus on the elements of the current node to discriminate them more finely. In this way, the user can more easily focus on a subset of elements. By introducing a command, for example a right double click on the mouse, the user can navigate alternately from one tree to the other displayed tree.

Claims (3)

REVENDICATIONS1. Method of navigation within a navigation tree created from data relating to elements, these data being associated with attributes according to classification or variable elements, characterized in that it comprises: a selection step by an operator of classification elements and attributes associated with classification elements, - a step of calculating a first navigation tree from the objects selected by the user and from a navigation strategy allowing to arranging the levels of the navigation tree, - a step of displaying on a screen of a terminal of said first tree by displaying nodes associated with a level and an attribute value, the links between the displayed nodes respecting the ordered levels of the first navigation tree, - a step of executing navigation commands for selecting a node in the first tree and displaying information associated with the selected node. 2. A method of navigation within a tree according to claim 1, characterized in that it comprises a step of comparing a value of an indicator determined from at least one attribute and at least one threshold, and a step of displaying a graphical distinction for each node of the first variable tree according to the difference between this value of a given attribute and at least this threshold. 3. A method of navigation within a tree according to claim 2, characterized in that the value of the indicator is determined from a combination of attributes. 5. 6. 7. A method of navigation within a tree according to any one of the preceding claims, characterized in that it comprises a rotation control of the representation of the first navigation tree. A method of navigation within a tree according to any one of the preceding claims, characterized in that it comprises a step of selecting a node defining a subset of elements accessible from this node, and a step of calculating and displaying a new navigation tree by using the navigation strategy applied to this subset of elements. A method of navigation within a tree according to claim 5, characterized in that the first shaft and the new tree are displayed simultaneously, the user entering a command to navigate alternately from one tree to the other displayed tree. A method of navigation within a tree according to any one of the preceding claims, characterized in that the step of calculating the first navigation tree is performed at a remote server, the terminal transmitting the creation data from the tree to the server.