WO2002095622A2

WO2002095622A2 - Method and apparatus for the presentation of data from a database

Info

Publication number: WO2002095622A2
Application number: PCT/HU2002/000046
Authority: WO
Inventors: Jozsef Dombi
Original assignee: Adixo Kutato Es Fejleszto Kft.
Priority date: 2001-05-23
Filing date: 2002-05-23
Publication date: 2002-11-28
Also published as: WO2002095622A3; HU0102165D0; AU2002258007A1

Abstract

The invention relates to a method for the presentation of data from a database. In the method, data elements are associated to visible geometrical objects, and the geometrical objects are illustrated on a graphical user interface. The geometrical objects associated to the data elements of the same data field are arranged in clusters, and the clusters are arranged along a substantially closed two-dimensional curve. The clusters are presented in a 3D visual representation, where the 3D visual representation creates for an observer the simultaneous perception of three substantially independent dimensions. A dimension of the geometrical objects is a function of the data value of a data element in a common data field of a selected record and at least one or more further data values of the data elements in that common data field. Those geometrical objects of a cluster, which are associated to data elements with a data value satisfying a predetermined selection criterion are presented in a simultaneously visually distinguished manner in the 3D visual representation.

Description

METHOD AND APPARATUS FOR THE PRESENTATION OF DATA FROM A

DATABASE

FIELD OF THE INVENTION

[0001] The invention relates relates to a method of visualization of data from a database, and more particularly, to a computer program for providing a graphic user interface based on the visualisation method.

BACKGROUND OF THE INVENTION

[0002] Electronic data is the essence of modern, technology-based society. Rapid advances in computer and network technology now provide access to an increasing amount of data. This increasing amount brings new problems in the manageability of information. The problem necessitates the assimilation of relevant information from large amounts of data, yet the size, scope, and complexity of data may surpass human cognitive abilities. It is emphatically important that information be extracted from the data and presented to a user in a manner perceptible and comprehensible by human cognitive abilities. The optimized three-dimensional visualization of data provided by the method of the present invention conveys global and local contextual information even for large sets of complex data. The present invention overcomes basic problems and limitations of both presenting large amount of data and navigating through the data visualization on a single display. The method of the present invention supports a users 's ability to perceive and absorb graphical display information more effectively.

[0003] Existing approaches have a number of significant drawbacks illustrated below. In each case the method of the invention offers a solution for the avoidance of these drawbacks.

[0004] Some conventional methods organize data into charts. The difficulty with charts is that they may be too large to be displayed in their entirety, and they may contain too much and too complicated numerical information. As a consequence, this type of arrangement of data may become confusing and incomprehensible to the user. The method of the invention alleviates this problem by visualizing all data - regardless of the size of the database - by one well-structured, easily comprehensible and navigable representation.

[0005] Other conventional methods require the user to leave a set of multiple windows and/or dialog boxes open, and switch between them in order to execute navigation and/or query operations. This method prevents the user from seeing and interpreting the operations and their consequences simultaneously by failing to provide a single integrated user interface. The method of the present invention does provide an integrated user interface, where navigation and/or query operations are available all in one place. [0006] Some further conventional methods often have to face the difficulty of visualizing both local and global aspects of data. An attempt to visualize global contextual information about large amount of data may distract a user from focusing on local aspects of data visualization. Conversely, displaying local aspects of data visualization may overwhelm the display of global contextual information. The special visualization technique included in the method of the invention allows the examination of local detail, while maintaining a global representation of the rest of the database by maximizing the integration of both local and global information into a single representation.

[0007] Again, other conventional methods provide little or no graphical visual feedback about the status of both navigation and/or query operations and query results. Rather, such methods merely return data results in a separate visual space; at best, in graphical format, such as bar charts, pie charts, specialized distribution functions, etc. These means may be too limited for the presentation of data in their entirety, or the resulting displays may be overloaded and confusing. The visualization method of the invention surpasses the simple graphical representations just described, by visualizing the data in the form of three-dimensional, animated representations.

[0008] Some further conventional methods may require the application of a complicated command set or formula language for the execution of navigation and/or query operations on a database. This may cause the user to misunderstand, or to be entirely unable to extract information from the queried database. In other cases, the user has to type in separate commands for each operation, which may be tiring and bothersome. In contrast to such methods, the present invention makes navigation and/or query operations available as simple mouse actions on a user interface, hereby this method allows the data to be queried interactively.

[0009] Finally, in comparison with methods that visualize databases in three- dimensional space, the method of the invention does not require significantly more computation, and may even be simpler. The method avoids three-dimensional occlusion, alleviates navigation and manipulation problems, and allows the display of any database, containing a practically infinite amount of information.

SUMMARY OF THE INVENTION

[0010] In an embodiment of the first aspect of the present invention, there is provided a method for the presentation of data from a database. The database comprises multiple records with at least two common data field. The database comprises multiple data elements in the common data fields. In the method, the data elements from the common data fields are associated to visible geometrical objects, and the geometrical objects are illustrated on a graphical user interface. The method includes the following steps:

[0011] The geometrical objects, which are associated to the data elements of the same data field, are arranged in clusters. These clusters are arranged along a substantially closed two-dimensional curve. The geometrical objects arranged in the clusters are presented in a 3D visual representation. The 3D visual representation creates for an observer the simultaneous perception of three substantially independent dimensions. The dimensions of the presented geometrical objects are selected so that a first dimension of a geometrical object is a function of the data value of a data element in a common data field of a selected record. Beside, the this first dimension of a geometrical object is also a function of at least one or more further data values of the data elements in that common data field, in which the further data values are taken from records different from the selected record. According to the method, those geometrical objects of a cluster, which are associated to data elements with a data value satisfying a predetermined selection criterion are presented in a simultaneously visually distinguished manner in the 3D visual representation. [0012] The term "simultaneously visually distinguished" means that the affected geometrical objects are illustrated in the 3D visual representation in a manner where an observer would notice that these geometrical elements are distinguished from the rest of the presented geometrical elements, and and this distinguishment appears simultaneously at the affected geometrical objects. The distinguishment may take the form of a striking colour, or a special position, or a general change of the appearance of the affected geometrical objects, or any other form of enhanced presentation which is readily apparent for an observer, and which is capable of drawing the attention of the observer to the affected geometrical elements. [0013] By assigning a geometrial object to a data element, even digital data values are transformed into an analog form. The transformation of data stored in digital form into analog form is analogous to how human understanding works. Digital form of data means primarily number-based form of data, while analog form means a form represented by variable measurable physical quantities. Analog data is more comprehensible to human understanding, therefore the transformation from digital into analog form is often necessary. The method of the invention, therefore can be applied to support human thinking and intuition in querying data.

[0014] The method enhances the presentation of data which are somehow interrelated. The study of such interrelated data is complicated with known forms of presentation, particularly if a large number of data fields needs to be studied. The invention gives an intuitive visual support to grasp interrelations between large amounts of data.

[0015] The method of the invention is particularly suitable for the visualization of large databases, however it may also be applied to smaller databases comprising at least one database table with multiple data records, and at least two data fields that contain multiple data values. Data fields may be of several types including numeric or discrete, out of which at least one numeric or one discrete data field is necessary together with an identifier field in order that the visualization of the information included in them be meaningful. An identifier data field contains item identifiers (names, for example) of the individual data elements in the database. Numeric and discrete data fields contain data values that describe the individual data elements. Multiple data values may have the same numeric or string value.

[0016] For easy distinction, the two types of data fields - numeric and discrete - are represented by different geometrical objects. Three-dimensional spirals are rendered to numeric fields and three-dimensional cylinders represent discrete fields. These cylinders can further be dissected to display the categories of the corresponding fields also as cylinders. The spirals show the values of the data elements by the distance from the bottom of the flat surface, while the height of the three-dimensional cylinders conveys information about the discrete fields and its categories. [0017] As a result, visualization according to the present invention allows the user to examine local details, while maintaining a global representation of the database. Further, the visualization method of the invention allows the animation of the geometrical objects. Animation is a sequence of representations being presented at a sufficient speed to produce the perception of continuous motion of the three-dimensional geometrical objects. The preferred form of animation of the geometrical objects is rotation. Rotation may apply to the circular plates of individual geometrical objects, groups of geometrical objects, and/or the plate comprising the entirety of geometrical objects. Rotation of the circular plates may happen independently of each other, or simultaneously with each other.

[0018] The direction of rotation depends on the current location of the geometrical object(s) under examination. The objects can rotate clockwise or counterclockwise depending on the position of the object selected to be viewed. The objects move synchronously together if an object representing a value is selected and independently if a category of a discrete field is selected to be examined. The rotation lasts till the selected object will be in the front-position. Due to the rotation, some geometrical objects and/or some parts of the geometrical objects are perceptible in the central area of the foreground, while others are perceptible in the background. However, the complete database remains on the display, which supports the perception of the interrelation between the data elements. This is particularly useful when small, portable computers with a relatively small screen are used, such as PDAs (personal digital assistants). [0019] Changing the position of the circular plates by rotation allows the user to extract information quickly and easily. An item of interest, currently being hidden in the background can be rotated into view. Examining the position of items in the sequence of others helps reveal hidden facts, correlations, similarities and exceptions. The animation of the geometrical objects supports the examination and selection of different data fields and/or different data values of the data fields.

[0020] A significant advantage of the method is that it can be used without any kind of formal training. Both navigation and query operations are available as simple, direct manipulation actions on iconographic representations of those operations, and do not require the learning of a command set or formula language. With other words, data can be interactively navigated and queried with simple mouse actions.

[0021] The invention is particularly useful for the analysis of large databases, such as a financial database of a company. The invention can effectually be applied in various areas by both non-technical users and experts, including business professionals, engineers, lawyers, scientists, medical professionals, students, as well as in the advertising, sales & marketing, and PR sectors.

BRIEF DESCRIPTION OF DRAWINGS

[0022] The invention will be now described with reference to the enclosed drawings, where

Fig. 1 A shows a possible layout of the graphic user interface used in the method,

Fig. IB is an explanatory figure indicating the functions of the various parts of the graphic user interface shown in Fig. 1 A, Fig. 2 is a schematic flow chart showing the general steps of the main routine for navigation operations on a user interface.

Fig. 3 is a schematic flow chart showing the general steps of the main routine for query operations on a user interface. Fig. 4 is a schematic perspective view of a cluster of geometrical objects associated to numerical data values, Fig. 5A is a schematic perspective view of a cluster of geometrical objects associated data values representing categories, Fig. 5B is a top view of the cluster of Fig. 5A

Fig. 6 is an example display view of the visualization, in the same layout form as illustrated by the schematic diagram in Fig 1,

Fig. 7 is an example display view of how the data being displayed appears as a single container of cylinder shape, that can further be dissected to show its content in more detail. Fig. 8 is an example display view of the data field listing area, including a textual description of the visualized, graphically represented data in the form of a hierarchical tree structure, Fig. 9 is an example display view of the data element listing area, including a list of data elements, the use of the option for marking data elements, Fig. 10 is an image of a display area containing buttons, which are used to control the contents of the display areas in the different windows,

Fig. 11 is an enlarged part of Fig. 6, illustrating a display of a 3D graphical representation of a database which contains numeric and discrete data fields, Fig. 12 is an illustration of a cluster of geometrical objects, resembling a spiral, where the height of the geometrical objects indicate the values of the data elements,

Fig. 13 represents the display of a specific data field in the form of a container of cylinder shape, Fig. 14 represents the data field shown in Fig. 13, where its container is open to show categories in the data field, each category being represented by a cylinder,

Fig. 15 is an illustration of a selection operation on the displayed data, showing also the markers already assigned to certain geometrical objects in the clusters, Fig. 16 is an illustration of a filtering operation on a discrete data field in the query window, where the query construction area is also shown, including the iconographic representation of query operations and of options for query result visualization, Fig. 17 is an illustration of a filtering operation on a numeric field in the query window. Fig. 18 illustrates the result visualization following a merging operation performed on two fields. Fig. 19 shows the query result representation on all the displayed and analyzed data,

Figs. 20 to 22 represent successive steps of rotating all geometrical objects representing the data, Fig. 23 represents a step of rotation in which only geometrical objects associated to one data field have been rotated. Fig. 24 represents a step of vertical rotation in which the position of objects on the flat surface horizontally remain the same as in Fig. 23 but vertically changes to display them from another viewpoint, Fig. 25 shows an example computer system for implementing the data visualization method.

DETAILED DESCRIPTION OF THE INVENTION BASIC CONCEPTS, TERMINOLOGY [0023] The term "data" technically refers herein to raw facts and figures that indicate or include information that is transitory or is being stored or transmitted. In general usage, the terms data and information are used synonymously. However, data can be any form of information whether in paper or electronic form. In electronic form, data refers to files and databases, text documents, images and digitally encoded voice and video. Data could exist as electromagnetic or other transmitted signals or as signals stored in electronic, magnetic, or other forms.

[0024] "Database" is a general term including the entirety of data with an organised structure that describes a specific area of knowledge. The organised structure of the database means that the database contains identifiers and indexes, by which data may be retrieved from the database according to certain selection criteria. By way of an example, a database of business companies may contain the entirety of data describing a set of business companies, including financial data, personnel data, administrative data, etc. A database contains at least one database table, or it may be the structured connection of database tables. [0025] The term "database table" refers to an orderly arrangement of data, especially one in which the data is arranged in rows and columns.

[0026] The data elements associated to a single data element of an identifier field are commonly referred to as a "record". A record is typically one row of a database table. A record typically includes data elements in the related fields about a data element or activity. [0027] A "data element" is the smallest single unit of the database stored in a field. A single data element can be one company in a database of business companies, or one type of car in a database of cars, or a value of a company stock at a given time, for example.

[0028] One column of the database table is referred to as a "data field" or "field". It is a physical unit of data. Data fields may be of several types including identifier, numeric, and discrete. A data field includes a number of data elements that describes and provides information about the data element or activity. The actual value of a data element is referred to as "data value" or "values".

[0029] An "identifier data field" contains identifiers of the individual data elements of the database. Data element identifiers may be names, like names of companies or types of cars, numbers, like post codes, or any other natural language, string or number identifiers. An identifier data field is typically the first column of a database table. Further columns of a database table are typically formed by numeric and/or discrete data fields. Numeric and discrete data fields contain data values that describe the individual data elements. As an example, the database may contain a data element which carries the information of the headquarter of a company. In that case, the data value of that data element is a city name, like "Seattle". Or, a data element may represent the annual income of a company in a given year. In that case, the data value of the data element will be a numeric value, like "1.650.340.000". The dimension of that value will be defined in another data field, where the data value of the elements in that field could be "US dollars" or "EURO".

[0030] A "numeric field" typically comprises numerical information about the data elements in the form of numeric values, i.e. numbers or quantities. A numeric field may contain the market values for a database of business companies, or the fuel consumption for a car database, for example. In such a case, the actual market value of a specific company, or the actual fuel consumption of a specific car is described by numeric values.

[0031] A "discrete field" typically contains information that categorizes the data elements. This information includes alternative, mostly non-numerical, or only a few numeric values in a form called categories. For example, a company may belong to the metal manufacturing industry group or the motion picture production industry group, but it is unlikely that it belongs to both. In a car database, the vehicles can be categorized by the type of fuel they consume; for example, a car can belong to either the category diesel or gasoline 98 or 95, respectively.

[0032] In the specification, numeric and discrete data fields are often referred to as "data attributes" or "attributes" when referred to in general. In individual references, numeric data fields are often referred to as "numeric attributes", while discrete data fields are often referred to as "discrete attributes". [0033] The data value that a numeric attribute can have is referred to hereinafter as "numeric value", while the data value that a discrete attribute can have is referred to hereinafter as a "discrete category".

[0034] "Structure" refers to the arrangement of underlying relations and hierarchy of data elements, data attributes, and data values, and to the visual representation thereof, where all data-related information is presented on a single display.

[0035] In the following, we refer to Figs. 1 and Figs. 4 to 24, which illustrate an embodiment of the method. Figs. 4 to 24 illustrate images on a computer display, or parts of such a display. The display shows a graphic user interface, which partly displays data of a database, and partly facilitates various operations with the data in the database. [0036] Fig. 1 shows schematically the graphic user interface (shortly GUI) displayed on the display of a computer. The GUI 10 of Fig 1 is shown in an optional arrangement. The upper left portion constitutes what is hereinafter referred to as data field listing area 12 (see also Fig. 8) The lower left portion contains is the display control area 14, and it contains animated buttons to control the display (see also Fig.10). The upper central portion is used primarily to show a more complex set of data from a database, and it is hereinafter referred to as the visualization area 16 of the GUI 10 (see also Fig. 11.). The visualisation area 16 includes a three-dimensional graphical representation of an entire data structure, according to the principles explained below. [0037] The upper right portion of the GUI 10 is hereinafter referred to as the query construction area 18 (see also Figs 16-18). The bottom right portion of the GUI 10 next to the control area 14 is hereinafter referred to as the data element listing area 20 (see also Fig. 9).

[0038] Fig. 2. illustrates the main steps in the routine run by the computer which controls the contents on the GUI 10. In a first step 110, the data are retrieved from a database, and the retrieved data are displayed on the GUI 10 in the form of geometrical objects, as explained later. If the user wishes to change certain aspects of the presented image, - particularly the viewing angle on the shown geometrical structures - the user indicates it towards the computer in a step 120. The computer calculates the necessary changes in the display, and in steps 130 and 140 calculates the images of the geometrical objects according to the requested new aspects. Finally, the new display is shown in step 150. These general steps 110-150 to present data according to user inputs are known in the art, and need no further explanation.

[0039] Similarly, Fig 3. illustrates the main steps in the routine run by the computer when query operations are performed on the displayed data, and the results of a query operation must be presented. In a first step 210, the data are retrieved from a database, and the retrieved data are displayed on the GUI 10 in the form of geometrical objects. If the user wishes to perform certain operations depending of the data contents - for example the analysis of data in a selected data field - the user indicates it towards the computer in a step 220. In a step 230, filtering conditions are sent to the computer, for example by mouse input (as indicated with 230A) or through iconographic interface (as indicated with 230B). Of course, other inputs, as with a keyboard, are also possible. The results of the query operation are performed in step 240, and the results are visualised in step 250. Again, the general steps 210-250 to present the results of a query operation according to user inputs are known in the art, and need no further explanation.

[0040] The present invention concerns the method with which the data of a database are presented visually for a user on a GUI 10. The GUI 10 may be displayed by a conventional display, such as the display 310 of the computer 300 shown in Fig. 25. The method is applicable for all databases which comprise multiple records with at least two common data field. Typically, one of these data fields would be an identifier field. The method is mostly useful where the database comprises a large number of data elements in the common data fields. For example, the database could contain the main financial data of the first thousand companies listed on a stock exchange.

[0041] In the method, the data elements from the common data fields are associated to visible geometrical objects. Not the data values of the data elements themselves, but the associated geometrical objects are illustrated on the graphic user interface 10. The principles of associating the geometical objects to the data elements are explained with reference to Figs. 4, 5 A and 5B, and Fig. 11.

[0042] Multiple data values, included in the data fields, may have the same numeric or string value.

[0043] The two types of data fields are represented by different geometrical objects.

[0044] Fig. 4 illustrates the representation of numeric values of a seleced data field. Here, the geometrical object associated to a date element is a vertical bar 32. These vertical bars are arranged in a cluster along a closed curve, namely the circumference of the circular plate 30. A dimension of the geometrical object is a function of the data value of the data element which is associated to the geometrical object. For example, as shown in Fig. 4, the height h of the bar 32 is a function of the associated data element, for example revenue figure for company A. Similarly, the height h' of bar 32' and the height h" of bar 32" are a function of the revenue figure for companies B and C, respectively. [0045] However, the heights of the bars 32,32' and 32" are not only a function of the directly associated data element, but also a function of at least one or more further data values of the data elements in that common data field, in which the further data values are taken from records different from the selected (associated) record (the selected or associated record of a geometrical object is that record which contains the data element which is associated to the geometrical object in question.) With other words, the values of the heigths of the bars are determined also taking into account the data values of other data elements in that field. In this manner, the presented geometrical objects convey information not only about the data value of the directly associated data element, but also about the relation of their data value to the other data values in that field.

[0046] As an example, the dimension of a geometrical object associated to a data element in a data field of a selected record is also a function of the number of those data element in the same data field of other records, which data elements have a higher or lower data value as the value of the data element of the selected record. With other words, the dimension of the geometrical object, i. e. the height h of a bar 32 associated to a data element will be determined on the basis of the order between the data values in the data field. With the given example, the height h,h' and h" of the bars 32, 32' and 32", respectively, may not reflect directly the the value of the revenue, but more the order between the values of the revenues. Further, the geometrical elements are arranged along the closed curve not according to their order in the database, but according to the value of the height of the bar. This means that the geometrical objects associated to the data elements of a data field are presented in monotone order along the closed curve, according to the value of a dimension of the geometrical objects. As a result of this monotone order between the geometrical objects, the clustes of the geometrical elements associated to data elements of numeric fields appear as three-dimensional spirals.

[0047] These spirals can be imagined as similar to distribution functions wound up in space. They represent a sequence where the pieces of information about the data set follow each other according to their value within the data field. The sequence is increasing, thus the higher the data value, the higher place it occupies in the sequence. A data element is presented as a bar on the surface of the spiral. With large number of data elements in a data field, the bars practically have a zero thickness, and they form a continuous barrel surface of the spiral.

[0048] In Figure 12, for example, the spiral presents the market values of companies. The market value of a company is indicated as a bar on the surface of the spiral, where the height of the bar is proportional to the market value of the company. If this spiral is viewed as a two-dimensional distribution function, it can be seen that the horizontal axis goes from the cheapest to the most expensive company, while on the vertical axis 0% stands for the lowest and 100% for the highest value. With other words, the dimensions of the geometrical objects are scaled so that the dimension of a geometrical object associated to a data element in a data field of a selected record is a function of the average or highest or lowest value of the data elements in the same data field of all records.

[0049] If the database contains discrete fields, in their display it may make sense to separate the data sets with identical values from each other within one field. It is noted that the discrete categories may be expressed by both numeric or text values. [0050] In the preferred embodiment, if the database comprises multiple record sets, which have a same data value in a common data field, a geometrical object is associated to a record set, and the geometrical objects associated to the record sets are arranged along a substantially closed curve. This is illustrated in Figs 5 A and 5B. The geometrical objects associated to the record sets with the same value in the common data field are cylinders 40ι-40s with a given height h. These geometrical objects arranged along the closed curve (see circle 42 in Fig. 5B) also constitute a well-defined cluster of geometrical objects. The record sets with the same value could be companies, which are categorised according to their activities. As illustrated in Figure 7 and 13, the discrete field "Industry group" taken from a financial database appears as a single cylinder on the display. Opening the cylinder reveals that there are five categories within the field each represented by one separate cylinder (see Figure 14). The height of the cylinders stand for the distribution of the data sets, showing the percentage of data sets belonging to one discrete category. Figure 14 shows that Banks have 26.20% distribution within the analyzed "Industry group". With other words, also in this case a dimension of a geometrical object is a function of the data value of a data element in a common data field of a selected record. For example, in Fig. 5A, the presented cluster of geometrical objects is the cluster of cylinders 40ι-40s. The presented data field is "Company activity". The data elements of the records are associated to a geometrical object, namely one of the cylinders 40ι-40₅. However, in the case of a discrete field, multiple data elements are associated to the same geometrical object, while in the example of Fig. 4, the different data elements in the data field (Company revenue) were each associated to a different geometrical object. But also in the case of a discrete field, the height of a cylinder 40 depends not only on the value of the associated data element, but also at least one or more further data values of the data elements in that common data field, in which the further data values are taken from records different from the selected record. In this manner, a quick glance on the cylinders reveal the relation between the data elements in the discrete data field.

[0051] For example, in Fig. 5A, the height h of the cylinders 40ι-40₅. reflect the percentual portion of the data elements falling in a certain catagory, relative to the total number of all data elements. Simply looking at the display, it is immediately apparent that the number of companies falling in that category which is associated to cylinder 40₅ in Fig. 5 A outnumber all other company categories. In this example, the geometrical objects associated to data elements of data fields with non-numerical data and with identical data values in multiple records are presented as a distribution, where a dimension of the geometrical objects reflect the relative number of records having identical data values in a common data field.

[0052] Discrete fields are therefore represented not only as single objects of cylinder shape, but these objects can also be dissected into groups of three-dimensional cylinders that represent the data sets, i.e., the categories in the discrete field. Categories within a data field may be regared as identifiers for selected record sets. In this manner, a first dimension of a geometrical object associated to a selected record set is a function of the number of records in the selected record set and the number of records in the remaining record sets.

[0053] The most significant advantage of the invention is the provision of a graphical data presentation, in which the essential features and properties of a database are highlighted, presenting all data of the database on a single screen, and providing a comprehensive structure for the presented data. This structure shows not only the data values themselves, but also the interrelation between data elements and data sets within the database.

[0054] As explained above, the geometrical objects associated to the data elements of the same data field are arranged in clusters.

[0055] The relation between the various data fields becomes apparent for the observer by arranging the clusters along a substantially closed two-dimensional curve, and presenting the clusters in a 3D visual representation, where the 3D visual representation creates for an observer the simultaneous perception of three substantially independent dimensions. As illustrated in Fig. 11, a dimension of the geometrical objects and the two dimensions of the two-dimensional curves are each associated to one of the three substantially independent dimensions of the 3D visual representation. As seen in Fig. 11, the height of the cylinders and the bars of the spirals is taken along the z coordinate, while the closed curve (a circle 50) lies in the plane X-Y. It is to be noted that within the clusters the geometrical objects are arranged along smaller circles, such as the circle 42 in Fig. 5B, while the clusters of the objects themselves are arranged within a larger circle 50.

[0056] The interrelation between data fields and data elements is visually highlighted for the user. This is achieved by the method by presenting those geometrical objects of a cluster, which are associated to data elements with a data value satisfying a predetermined selection criterion, in a simultaneously visually distinguished manner in the 3D visual representation. With other words, certain selected data elements, which have some common feature, and which otherwise could be in quite different places in the database, are visually distinguished, and this visual distinguishment appears simultaneously for the user. By the term "simultaneously visually distinguished " it is meant that the selected geometrical objects are marked in visually perceptible manner, and this marking appears simultaneously for the user, who also perceives that the marked geometrical objects have a common connection. The visual marking or distinguishment may take several forms. For example, a common colour of the marked objects may be strikingly different from the colour of the other geometrical objects, or their position relative to the user or relative to the closed curve of the clusters could be similar. [0057] In this manner, distinguished records or data sets comprising multiple records within the database may be exposed, while maintaining their position between the other records of the database, and their relation to the other data is also conveyed to the user. For example, in a typical application the geometrical objects associated to a set of records with equal data values in a common data field are presented in a simultaneously visually distinguished manner. As a specific example, all companies belonging to the "Software sector" may be exposed, while also their financial data, such as market value or revenue is also shown together with the data values of the financial data of other companies, and, more importantly, their relation to the data values of other companies [0058] The method may be applied to, and is particularly suitable for the visualization of large databases comprising multiple database tables that contain multiple data records, and multiple data fields with multiple data values. However, the method of the invention may also be applied to smaller databases. The minimal conditions, when the application of the invention is meaningful, suppose the existence of a database comprising at least one database table that contains multiple data records, and at least one, but preferably two data fields with multiple data values. One of the two data fields should be identifier, while the other field could be optionally numeric or discrete. Data fields, describing the data sets, are displayed by visible geometrical objects, (spirals and cylinders), where a specific parameter of the geometrical object, (height), represents the value the data field takes up at a certain point. These geometrical objects, - associated with the data fields, are arranged in a predetermined order on a circular plate, and are presented as a three-dimensional, rotating image. In the presented embodiment, a geometrical object or a cluster of geometrical objects have a similar layout as the substantially closed curve, around which the clusters of the objects themselves are arranged. In this manner, multiple levels of data structures may be presented on a single screen. The clusters of geometrical objects may be organised into larger groups, which again may be displayed arranged along a closed curve. Alternatively, if a data set containing several data elements are associated to a geometrical object, an internal structure of such a data set may be exposed using the same display format, without the necessity to visually rearrange the originally displayed and not affected geometrical elements. The number of displayed structure levels is only limited by the resolution of the used computer display.

[0059] In the shown embodiments, the substantially closed curve is a circle, but other substantially closed and - preferably - more or less symmetric curves are also useful. The use of a closed curve supports the visual comprehension that the complete database or all data elements of a data field are displayed, and also ensures that the observer constantly sees the spatial relationship between the clusters of geometrical objects, and through the spatial relationship perceives the logical connection between data fields and other data sets. A closed curve has no beginning or an end. Therefore, the use of a substantially closed curve for arranging the visual representation of the data also conveys to the user that there are no "first" or "last" data elements, and there is no special distinction between the data elements which would indicate their original position in a database table, but it is their relation to the other data elements what really counts. This is particularly true for discrete categories. Preferably, the geometrical objects associated to the data elemens are lines or circular columns (cylinder or bar). Other objects, e. g. rectangular columns are also applicable. A line may be considered as a circular column with zero diameter.

[0060] A typical application where a selection of the data elements takes place, is a query operation on the database. In the method, those geometrical objects are presented simultaneously visually distinguished , which are associated to data elements resulting from a query operation. Fig. 15 to Fig. 19 are example display views of query operations, showing also the query construction area according to the present invention, including partial three-dimensional graphical representation of data from a database and iconographic representation of query operations and of options for query result visualization.

[0061] Another typical application is where the geometrical objects associated to data elements of different data fields of the same record are presented in a simultaneously visually distinguished manner. Because the data elements of a record are shown together with all data of the database, the position of a data element may be compared quickly with the data elements in the same data field. For example, the market value, the revenue, market share etc. and other financial indicators of a selected company are displayed on a single display, immediately showing the position of the company according to multiple criteria.

[0062] In a further typical application is the geometrical objects associated to a set of records with equal data values in a common data field are presented in a simultaneously visually distinguished manner. For example, all companies belonging to a certain industry sector may be highlighted simultaneously, in all affected data fields.

[0063] In order to improve the overall visual appearance of the geometrical objects, and to enhance the perception of the data, geometrical objects associated with values of the selected data set may be marked by a color different from the color of the geometrical objects associated to other data sets. Such selected data can be seen in Fig. 15, with highlighted geometrical objects. The highlighted geometrical objects are also identified by text information (3com, app). The natural language description assigned to the elements on display is an additional means to help visual perception of the data elements. Selected data elements may be presented with their distinctive descriptions that can consist of their names or any other identifier text, and also their values in the field being currently under examination. This support together with the other visualization helps a user identify the elements and quickly discover useful information. Such textual information may be displayed on various parts of the screen, typically in the top or bottom parts of the GUI 10.

[0064] In a further improvement, the simultaneously visually distinguished presentation of the geometrical objects is animated. If the simultaneously visually distinguished presentation or the marking of a geometrical object involves the change of the position or size of the geometrical object, the change is made continuously, so that the user perceives a continuous movement of the geometrical object from the starting position to the end position. In this manner the observer does not lose the perception of the complete data structure, and the interrelation between the data elements and data fields. In a preferred form, animation is realised in the form of rotation. As explained above, the geometrical objects (spirals and groups of cylinders) are organized in smaller circles on a larger circle, which functions as a circular "plate". This circular plate rotates. Each group of cylinders may also be regarded to be on a separate smaller circular plate, which is also able to rotate independently of the larger circular plate. Spirals form a circular plate themselves and are also able to rotate independently of the large circular plate. With other words, the animation of a cluster of geometrical objects comprises rotation of the clusters around a principal axis of the affected cluster. Alternatively, the animation involves rotation of the complete set of clusters around a principal axis of the closed curve, i. e. that closed curve around which multiple clusters are arranged.

[0065] Changing the position of the circular plates allows the user to extract information quickly and easily. Animation enhances the examination and the selection of different data values and data fields. Figure 20 to Figure 24 represent how navigation can be performed on the data by rotating either all data or only portions of it. Comparing Figs. 20 to 24, it will be apparent that the same data structure is presented, but from different viewpoints. For example, in Fig. 20, a spiral representing the numeric field "Earnings jper_share_1995_actual" is in the foreground, with the record associated to the company "TANDEM COMPUTERS" being closest to the observer. At the bottom of the display area, the exact value of the selected data element is also shown (0.91), with an indication of the percentage (32.4%) of this value relative to the highest value. However, from the position of the spiral it is also apparent that the company "TANDEM COMPUTERS" is somewhere in the lower third of all companies in this data field. In Fig. 21, another spiral representing the numeric field "Assets_12_Months_19951" is in the foreground, with the geometrical element associated to the record of the company "PACIFICARE HEALTH SYSTEMS" being closest to the observer. In this navigation sample, rotation has been executed in a counter-clockwise direction on the circular flat surface where the objects, representing different data elements are located. The flat surface is capable of being rotated either in a clockwise or a counter-clockwise direction, and it is also possible to be moved in a vertical direction showing the objects from different upside- down viewpoints. The objects representing elements of a single field can also be moved independently of the whole data. Comparing Fig 22 and Fig. 23, only the cluster in the foreground have been rotated (associated to the data field "Sales_sign"). In Fig. 22, the category "z" is in closest to the observer, and the colouring on the top of the cylinder indiates that the centrally displayed numeric information ("1.48% [0.00%] ) relates to this category. By comparison, the cylinder associated to the category "N/A" is closest to the observer in Fig. 23.

[0066] A further advantageous feature of the method that different querying possibilities are available to the user as simple mouse actions. There is no need to learn any complicated command set or query language to perform them so any non-technical user can easily become familiar with the procedure. An interactive graphic user interface is provided to initiate a query operation on the database.

[0067] Query operations are executed in different query windows on the chosen data field. Successive operations can be performed on each displayed field, and the program will calculate the query result according to all the executed actions. In a preferred embodiment, an input value of the query operation is selected by selecting a geometrical object in the 3D visual representation, where the selected geometrical object is associated to a data element with a data value which is used as the input value of the query operation. For example, by clicking with the mouse on the geometrical object associated to the sales volume of a company, the value of this sales volume may be selected to be a threshold value, and the query operation finds the share values of companies having a sales volume above the threshold value.

[0068] Typically, the following query operations are useful with the method, as described hereinafter. SELECTION

[0069] Selection is an additional feature of the method of the invention, which is designed for the detailed examination of data elements, values, and data fields. Selection can be conducted in different ways, which permits the examination from different aspects [0070] If data sets are selected, the animated geometrical object will be updated to display information about the values describing the data set within all data fields.

[0071] Data fields can also be selected, in which case data sets and values included in the field will be displayed.

[0072] In the third case, a value can be selected, where the data sets containing the selected value will be shown to the user. FILTERING AND VISUAL QUERY [0073] Filtering, according to the present invention is a visual query operation that may be applied to the individual data fields. The herein preferred meaning of filtering is inquiring information about data under restrictive conditions. With other words, an arbitrary range of data values may be filtered by defining restrictive conditions on the entire set of values, and the database may be queried to show data that satisfy the defined conditions. As a consequence, the data elements described by data values that satisfy the filtered conditions will be presented by graphical representations. This is illustrated in Fig. 16 and Fig. 17, where the results of the filtering are shown by colouring, indicated by the dark top of the cylinders on Fig. 16, and by the differently coloured bars of the spiral of Fig. 17. In this case, the selection criterion for selecting those data elements which are to be presented in a simultaneously visually distinguished manner are defined as a filtering condition of the data values of the data elements. The filtering conditions may be formulated in a number of ways. A specific filtering condition selects data elements of different data fields of the same record. Another specific filtering condition selects data elements of those records which have identical data values in a common data field. Filtering conditions may be combined. Thereby the user can examine data that answer multiple filtering conditions, excluding masses of unnecessary information from the procedure. This means of the method helps improve the speed and efficiency of the analyses on databases containing a huge amount of data. MERGING

[0074] Merging is another type of query operation that means, by preference, a kind of summation or intersection of two or more data fields. Merging can be completed on the bases of at least one discrete and one numeric data field. In a specific example, by merging data fields, the examination of the total real value of the data elements by the merged data fields may be realized. Further, merging allows the examination of the relation between the merged data fields and their influence on each other. Suppose the user needs to analyze a factory and make decisions on how to improve performance, productivity and suggest reductions on expenses. This procedure should include the examination of how profitable each separate factory unit is. By merging, the user not only can show the production of the individual units, but can also visually display the realized profit distributed to these units. The results of a merging operation are shown in Fig. 18, where the fact of the merging is indicated by the smaller disc on the top of the cylinders. The merged fields are textually indicated on the top of the display (Industry_group/Earnings_per_share_1995_Actual). The textual information at the bottom of the display indicate the category shown in the foreground (Fuel). In the present case, the geometrical elements associated to the resulting data set of the query were not marked by colouring (though such a marking would be also possible), but the data elements included in a resulting record set from a query operation are associated to new geometrical elements, and the new geometrical elements are displayed in the 3D representation. MARKING

[0075] In the proposed presentation method, marking may be considered as a kind of permanent selection. Marking is designed for the user to label the data element(s), which is(are) typically of special importance to the user. The marked data element(s) will be displayed in all portions and all positions of the three-dimensional graphical representations consequently. The preferred function of marking is to support comparison of the marked data element(s) with any further selected data element. An example of marking may be seen in Fig. 15 and 16. Marking is performed directly on the GUI appearing on the screen of a computer display unit. In the GUI, such as the GUI 10 shown in Fig. 1A and IB, the data is presented visually on a screen. [0076] In the method explained above, the data elements which are selected for synchronised presentation are determined by a selection function. This selection function requires an input value. Most conveniently, the synchronised presentation of data elements is initiated by marking on the screen a geometrical object. The marking of the geometrical object also identifies a data value, namely the data value of the data element which is associated to the geometrical object. In this manner the input value of the selection function may be chosen to be the data value of the data element associated to the marked geometrical object.

[0077] For example, in Fig. 19 and Fig. 20, the graphical representation of the geometrical objects, i. e. both the colour of the cylinders and colour of the bars on the spiral will change to indicate the items the values of which have satisfied the filtering condition. On cylinders, representing either discrete fields or the categories in the fields, the color of the surface may change direct proportion to the number of elements that have satisfied the filtering condition. In the case of a numeric field, the value representation of items on the spiral will change to indicate the items the values of which have satisfied the filtering condition. Categories of fields, created by merging a discrete field and one or more numeric fields have rings on top of their cylinders. With other words, the results of the query operation are presented in the 3D visual representation in a simultaneously visually distinguished manner by modifying the colour, position or size, or all of these, for those geometrical objects which are associated to the data values of the data sets included in the results of the query operation. It must be noted that in a preferred embodiment, not only the position of the selected geometrical objectes will be modified, but complete clusters of geometrical objectes are moved, so that the selected geometrical element will be in a specific position in the new position of the cluster.

[0078] A characteristic feature of the method is synchronicity. Synchronicity means that the selection and animation occur in a synchronized manner, i.e. data values of different data fields that describe a data set are presented in a synchronized manner, with other words, parallel or simultaneously in time. Synchronicity is also applies in all types of selection. The method visually emphasizes the data values, associated with the selection, in a way that is immediately noticeable to the user. For example, if a certain type of car is selected from the database, the data values describing this car will become highlighted on the geometrical objects associated with the respective data fields. The geometrical objects are turned to the user one after the other applying the rotating circular plates, so that the values describing the selected data set are visible. This allows quick comparison of data sets. [0079] Synchronicity also works in the case of a such geometrical object which are associated with a certain data field. Geometrical objects associated to data elements from those data sets which have identical values within a data field, are presented in a similar position relative to a viewing direction along the circular plate. Alternatively, the clusters or geometrical objects are presented in a similar position relative to the circumference of the circular plate. E.g., it is possible to select from a database of cars all cars that have the same engine power. Again, other geometrical objects, representing other data fields, e.g., price, cubic capacity, are turned so that the values of these fields for cars with the same engine power will be shown. This way, it can easily be judged how the selected engine power relates to those of other cars in the same engine category. [0080] An advantageous feature of the method is that it can be used without any kind of formal training, so query operations will be available to and can be conducted by non-technical users as well. Operations, including animation, selection, marking, filtering, etc., do not require to learn a complicated command set or formula language. Thus, data can be interactively queried with simple clicks of a mouse. The visual presentation of data helps human understanding and perception, therefore discovering useful business information will be much more easier by using the method of the present invention.

[0081] The method is described in terms of an example computer and data provision environment. Given the description herein, it would be obvious to one, skilled in the art, to implement the present invention in a computer program. Such a computer program is capable of running on known general-purpose or specialised computers. Such computers are normally equipped with a necessary program storage means to store the computer program. An examplary, general layout of such a computer 300 is shown in Fig. 25. The computer program may be stored on a suitable storage means of the computer, such as the hard disk drive 320. However, the computer program may be stored also on a separate data carrier, such as the removable storage means 340, with a known structure.

[0082] In one preferred example, the invention can be implemented as a software in an operating system, such as WindowsNT® or Linux®, with a required memory capacity, a graphics workstation. The displayed database may be handled by any known database management software, which is capable of performing the relatively quick retrieval of the data values in the database. It is not intended that the invention be limited to specific applications, operating systems or databases. In fact, after reading the above detailed description of the invention, it will become apparent to a person, skilled in the relevant art, how to implement the invention in alternative environments.

[0083] The present invention is particularly useful for the analysis of large databases, such as a financial database of a company. The presentation of such databases would normally necessitate the use of complicated charts containing large amount of numerical type of information. With the method of the invention it is possible to efficiently analyze the information hidden behind the numbers, spot the trends, and single out exceptions even in a complicated database. [0084] The invention can effectively be applied by business professionals, engineers, lawyers, scientists, medical professionals, students, as well as in the advertising, sales & marketing, and PR sectors. Numerous different databases can be analyzed to reveal information, business opportunities, possibilities for cooperation, meaningful relations etc., which could lead to new conclusions, well-based decisions, and discoveries in any of the above mentioned sectors. EXAMPLE:

[0085] Suppose that a business analyst, an efficiency expert, or an investor, etc. intends to execute a survey on companies in order to have a view of their financial and market positions, performance, etc. A further objective can be to analyze companies prior to making a decision whether or not one would be worthwhile to cooperate with or to invest in.

[0086] The example analysis is presented on a sample financial database that contains data of an industry group including companies in different sectors. This sample database, used both in the description and in its representation, is only for illustration purposes to the present invention and does not show real and current values.

[0087] The available fields that contain the attribute values of the companies under examination are listed in a tree form as shown in Figure 8, and are displayed in a separate window of the GUI 10. The complete database is illustrated with a container-like geometrical object shown in Fig. The database, i. e. the container of Fig. 7 may be dissected to reveal both its structure and content as seen in Figure 6. According to the types of fields, data is displayed in either spiral or cylinder shape forms.

[0088] A numeric field contains numerical data for an attribute of the data elements in each sector of the industry group such as "Market value", "Dividends", "Earnings Per Share", etc. These appear in the form of spirals on the circular plate of the whole database. A discrete field includes mostly non-numerical information or information consisting of only a few numeric values describing an attribute of the elements. These values form the categories of the discrete fields. Examples of these fields are "Computers", "Health care", "Banks", "Fuel", etc. Each field is represented by a cylinder, which becomes visible after dissection on the main circular plate surrounded by the circle 50, as shown in Figure 11.

[0089] To globally analyze the database, the data is displayed in the discovery window (visualisation area 16) as seen in Figure 6, where the category "Software/Services" from the field "Industry group" is active. Useful information can be acquired about its position and size on market by analyzing the height of its cylinder and comparing it to the other members of the group. Regarding the whole database, the height of its cylinder shows that there are sectors of similar and even smaller size on the market. However, there are also more significant sectors. Another feature of the method of the invention is the natural language description to the displayed element, which also helps the knowledge acquisition. The user can precisely see that the Software/Services sector has a 18.45% portion in the Industry group.

[0090] The user can also continue the analysis by taking the attribute values under microscope. As illustrated in Figure 12, a cluster of geometrical objects associated to data elements of a numeric field, in our example "Market Value" is represented as a spiral, with the values on the wall of the surface in ascending order. This arrangement helps the selection by easily filtering out either the highest or lowest values, or any in the sequence, respectively. Selections of data can be made according to the user's requirements in the query window; however, a selection by filtering a numeric field is shown in Figure 17. The filtering condition is shown by a color change.

[0091] A query operation by setting filtering conditions on a discrete field is shown in Figure 16, where "Industry group" is displayed in the query window, and the filtering condition is set to two categories in the field, i.e. for Computers and for Software/Services. It means that when calculating the result, elements in these two sectors will be included in the query result.

[0092] The query result is shown in the main window on all the objects representing the whole database by changing the color of the elements satisfying all the filtering conditions. The representation of it is shown in Figure 19, where the color of data values on the spirals changes if a value has met the filtering conditions, and the color on the surface of a cylinder representing a category of a discrete field changes in direct proportion to the number of elements within the category that have met the filtering conditions. This visual illustration of the result helps recognize the information resulted after one or more querying operations. On the spirals, the user can see how the values belonging to the selected elements distribute over the data sequence, and on the cylinders it is visible in the blink of an eye, what portion of the corresponding sector satisfied the requirements. [0093] To more thoroughly analyze the individual items, it is also possible to display them in a separate window on the GUI 10, for example in the query construction area 18. After executing the querying operations, the user can specify to display only those items in the visualisation area 16 that have satisfied all the conditions and belong to the selection. [0094] Merging on fields is one more querying operation the method is capable of executing. It is useful if not only the percentage values indicating the portions of categories, but also the real values in terms of a numeric field, i.e. a numeric attribute may be of special importance. For example, Figure 18 illustrates a merging operation, where the base discrete field is "Industry group", for which the values in the numeric field "Earning ' Per Share" has been distributed and the values of each category calculated. The textual information at the bottom shows that the Earnings Per Share value of the Fuel sector is 9.65% in the whole group. After the merging, the height of cylinders will show the total sum of each category calculated by the values in the numeric field.

[0095] Navigation through the data is another preferred feature of the present method. Navigation may be performed on the data either before or after querying it. It is useful, for example, if the user would like to successively see the attribute values of a chosen company. Navigation is straightforward, simply by rotating the object together or individually. These navigation possibilities help visually analyze the data and spot facts that would otherwise be hidden from the user.

Claims

CLAIMS:

1. A method for the presentation of data from a database, the database comprising multiple records with at least two common data field, and further the database comprising multiple data elements in the common data fields, in which the data elements from the common data fields are associated to visible geometrical objects, and the geometrical objects are illustrated on a graphical user interface, in which a, the geometrical objects associated to the data elements of the same data field are arranged in clusters, b, the clusters are arranged along a substantially closed two-dimensional curve, c, the geometrical objects arranged in the clusters are presented in a 3D visual representation, the 3D visual representation creating for an observer the simultaneous perception of three substantially independent dimensions, where

- a first dimension of a geometrical object is a function of the data value of a data element in a common data field of a selected record and at least one or more further data values of the data elements in that common data field, in which the further data values are taken from records different from the selected record, and further

- those geometrical objects of a cluster, which are associated to data elements with a data value satisfying a predetermined selection criterion are presented in a simultaneously visually distinguished manner in the 3D visual representation.

2. The method of claim 1, wherein the geometrical objects associated to data elements of different data fields of the same record are presented in a simultaneously visually distinguished manner.

3. The method of claims 1 or 2, wherein the geometrical objects associated to a set of records with equal data values in a common data field are presented in a simultaneously visually distinguished manner.

4. The method of any one of claims 1 to 3, wherein the simultaneously visually distinguished presentation of the geometrical objects is animated.

5. The method of any one of claims 1 to 4, wherein the data is presented visually on a screen, the data elements selected for synchronised presentation are determined by a selection function requiring an input value, and the synchronised presentation of data elements is initiated by marking on the screen a geometrical object, where the input value of the selection function is the data value of the data element associated to the marked geometrical object.

6. The method of any one of claims 1 to 5, wherein a first dimension of the geometrical objects and the two dimensions of the two-dimensional curves are each associated to one of the three substantially independent dimensions of the 3D visual representation.

7. The method of any one of claims 1 to 6, wherein a geometrical object or a cluster of geometrical objects have a similar layout as the substantially closed curve.

8. The method of any one of claims 1 to 7, wherein the substantially closed curve is a circle.

9. The method of any one of claims 1 to 8, wherein the geometrical object is a line or a column.

10. The method of any one of claims 1 to 9, wherein the database comprises multiple record sets, which have a same data value in a common data field, a geometrical object is associated to a record set, and the geometrical objects associated to the record sets are arranged along a substantially closed curve.

11. The method of claim 10, wherein a first dimension of a geometrical object associated to a selected record set is a function of the number of records in the selected record set and the number of records in the remaining record sets.

12. The method of claim 11, wherein geometrical objects associated to data elements of data fields with non-numerical data and with identical data values in multiple records are presented as a distribution, where a dimension of the geometrical objects reflect the relative number of records having identical data values in a common data field.

13. The method any one of claims 1 to 12, wherein a first dimension of a geometrical object associated to a data element in a data field of a selected record is a function of the number of those data element in the same data field of other records, which data elements have a higher or lower data value as the value of the data element of the selected record.

14. The method any one of claims 1 to 12, wherein a first dimension of a geometrical object associated to a data element in a data field of a selected record is a function of the average or highest or lowest value of the data elements in the same data field of all records.

15. The method of any one of claims 1 to 14, wherein the geometrical objects associated to the data elements of a data field- are presented in monotone order along the closed curve, according to the value of a dimension of the geometrical objects.

16. The method of any one of claims 4 to 15, wherein the animation of a cluster of geometrical objects comprises rotation of the clusters around a principal axis of the affected cluster or around a principal axis of the closed curve.

17. The method of any one of claims 1 to 16, wherein an interactive graphic user interface is provided to initiate a query operation on the database.

18. The method of claim 17, wherein an input value of the query operation is selected by selecting a geometrical object in the 3D visual representation, the selected geometrical object being associated to a data element with a data value which is used as the input value of the query operation.

19. The method of claim 17 or 18, wherein the results of the query operation are presented in the 3D visual representation in a simultaneously visually distinguished manner by modifying the colour and/or position and/or size of those geometrical objects which are associated to the data values of the data sets included in the results of the query operation.

20. The method of any one of claims 17 to 19, wherein the selection criteria of the data elements are defined as a filtering condition of the data values of the data elements.

21. The method of any one of claims 17 to 20, wherein a filtering condition selects data elements of different data fields of the same record.

22. The method of any one of claims 17 to 20, wherein a filtering condition selects data elements of those records which have identical data values in a common data field.

23. The method of any one of claims 17' to 22, wherein the data elements included in a resulting record set from a query operation are associated to new geometrical elements, and the new geometrical elements are displayed in the 3D representation.

24. A computer program product, comprising commands for implementing the method of any one of claims 1 to 23.

25. An apparatus, preferably a computer, comprising a memory storing a computer program product of claim 24, the computer program product comprising commands for implementing the method of any one of claims 1 to 23. A data carrier, storing a computer program product of claim 24, the computer program Q o. product comprising commands for implementing the method of any one of claims 1 to 23.