JP2016530650A - System and method for interactive visual analysis of multidimensional temporal data - Google Patents

Abstract

Multidimensional temporal data can provide insights into patterns, trends and correlations. Traditional 2D charts are widely used to support the work of domain analysts, but are limited to presenting large, complex data intuitively and cannot gain insight from further research. A visual analysis system and method is provided that incorporates the idea of a new visualization method to support interactive analysis of multidimensional temporal data. The system extends the capabilities of mapping technology by visualizing domain data based on 3D geometry enhanced by color, motion and sound. The system enables a compact and universal overview of large-scale data and deep drilling for further investigation. Customizable visualization can be adapted to different data models and applied to multiple domains. The system helps analysts interact directly with large-scale data to gain insights and make better decisions.

Description

  The present invention relates to computer-based systems and methods for interactive analysis of multidimensional temporal data, and more particularly to computer-based systems and methods for three-dimensional visualization of large scale data.

  Multidimensional and temporal data is very common in many business and scientific areas. Today, large-scale data is growing at an explosive speed everywhere. While the ability to collect and store new data is growing rapidly, the ability to analyze these data volumes is increasing at a much slower rate. Analyzing large and complex data is not easy, but is useful for gaining insight into patterns, trends and correlations. This means that there is great potential value in large and complex data.

  In the past, researchers have proposed many automatic analysis methods to process and simplify these data, such as principal component analysis, self-organizing maps and clustering algorithms. Another typical method for processing large-scale data is online analytical processing (OLAP) in which data is partially pre-aggregated in “cubes” and stored in a “data warehouse”. However, traditional analysis systems do not provide an effective and flexible way to present data or provide intuitive analysis results.

  Traditional visualization methods such as two-dimensional (2D) lines, bars and tables are simple and intuitive and are therefore widely used in our daily lives. However, they are limited in presenting large and complex data. For example, FIG. 1 shows a 10-year visualization of the Chinese mutual fund market at 100 and 101, identifying patterns and trends in a 2D table at 100, where each cell only contains numbers. That is difficult for analysts. The 2D bar at 101 is more suitable to show past and future trends. However, the 2D bar at 101 is not suitable for displaying additional detailed information, such as additional information about a particular type of investment trust.

  Other visualization methods such as scatter plots, heat maps, parallel coordinates, tree maps, internet maps, spiral graphs, density-based distribution maps, sphere base maps, theme rivers and time wheels have been proposed over the past decades Has been. These methods are useful in visualizing certain types of data. For example, a heat map can be applied to show financial information. Spiral graph is a visualization method for finding periodic patterns of influenza cases. Parallel coordinates are a common way to analyze multivariate data by visualizing high-dimensional geometry. The “cross-filter view” method can visualize and analyze multidimensional data. This method interactively represents a sequence of multi-dimensional set queries with cross-filtering data values across view pairs. Other proposed methods for visual analysis of temporal data are based on three main criteria: time, data and representation.

  In recent years, visual analysis (VA) has been introduced to represent large-scale multidimensional and temporal data with various visual encodings. Visual analysis can explain data patterns, trends and correlations in the shortest amount of time in the least amount of space. The advantages of visual analysis include the ease of design and the ability to analyze high complexity data. Visual analysis combines interactive visualization and effective analysis techniques for effective understanding, reasoning and decision-making based on very large and complex data sets. Visual analysis can be applied in many areas such as finance, health, geography, physics, security and so on.

  Visual analysis systems and tools can support domain analyst decision making and insight discovery through advanced visualization methods and system user-centric interactive operations. One popular commercial tool for visualizing business data is Microsoft Excel, which provides a standard visualization method for spreadsheet data (bars, columns, lines, circles, etc.). However, these visualization methods are constrained when the underlying data model is composed of complex ideas that need to be communicated with clarity, precision and efficiency.

  A wide range of companies ranging from specialized data discovery vendors such as Tableau, QlikTech and Spotfire to multinational companies such as IBM, Microsoft, Oracle and SAP, unique commercial visual analysis systems for analyzing a vast and growing amount of data Has been engaged in efforts to develop. Large software vendors tend to focus only on a few "standard" visualization techniques such as line graphs and tables that have limited capabilities in processing large, complex data. Existing toolkits for analyzing data are described in, for example, the Info Vis toolkit by SenchaLabs, Redwood City, Calif., The Open Source Foundation under the BSD license, SourceForge. Includes Protovis by Stanford Visualization Group under Prefuse by NET and BSD license.

  In academic terms, several VA systems have been developed to support the work of domain analysts. MobiVis, a visualization interface design innovation (VIDi) from UC Davis, is a system that visually analyzes mobile data by presenting social and spatial information in one heterogeneous network. VIS-STAMP by the Department of Geography at the University of South Carolina (Spatial Data Mining and Visual Analysis Laboratory) is a visual query system for spatiotemporal and multivariate patterns. VIS-STAMP supports different complex patterns and allows system users to focus on a specific pattern and examine a detailed view of the data through various interactions. WireVis by Bank of America and UNC Charlotte is a system that combines multiple visualization methods to analyze categorized time-varying data from financial transactions. Weijia Xu et al. Designed a system based on a treemap for analyzing large digital collections through interactive visualization. Hotmap is based on heat maps to visualize geographic data using the structure of the underlying data set to visualize it in its own space.

  Many researchers are involved in efforts in developing high-level models of VA system design. Specifically, Tamara Munzner characterizes tasks and data in problem domain vocabulary, abstracts them into operations and data types, designs visual encoding and interaction techniques, and efficiently executes the techniques We proposed a nested model for visual design and verification by four layers, which is to create an algorithm of Based on that model, Xiaoyu Wang et al. Proposed a two-stage framework for designing visual analysis systems in an organizational environment.

  However, the visual analysis tool described above provides a flat representation of vast amounts of data and cannot properly process multidimensional data. Existing tools have the difficulty of presenting time information on the data. These conventional approaches for temporal representation include, for example, heat maps or rectangular geometries. However, these visualizations are not intuitive and do not adequately convey the temporal elements of the data so that system users gain insights and trends in large and complex data sets.

The computer-implemented visual analysis system and method of the present invention provides system users with the ability to determine the occurrence of a dataset for a baseline standard presented as a global sphere. The global sphere can be segmented according to a predetermined rule for separating data sets for placement in the segmented portion of the global sphere. The present invention also includes receiving the input from one or more data feeds, including data feeds from one or more databases, and three dimensional according to a first predetermined rule for determining the radius of the global sphere. Creating a global sphere and displaying it on a display electronically connected to a computer. The global sphere serves as a baseline standard measure by which the occurrence of the data set is measured. The present invention also includes a step of segmenting a global sphere according to a longitude portion to define a first reference for placing each data point in one of the longitude segments; Segmenting a global sphere according to a latitude segment to define a second criterion for placing each data point on one of the data points. Thus, two criteria are used to determine the position of each data point with respect to the global sphere. According to the present invention, time T ₀ can represent the time when each data point is located at a surface position on a three-dimensional global sphere having a radius determined by a first predetermined rule. . Each data point is represented by a three-dimensional sphere having a radius determined by a second predetermined rule, and the radius for each data point is affected over time by at least a first performance criterion.

The system and method of the present invention tracks the occurrence of each data point from time T ₀ to time T _N according to information about each data point received from one or more data feeds or one or more databases, One of the following for the data points can be displayed on the display: (1) Determine the radius for the data points and first determine if the performance of the data points exceeds the global sphere baseline standard indicator In order to change the first performance criterion in this way, the length of the radius of the three-dimensional sphere associated with the data point is changed according to a second predetermined rule, and (2) the radius for the data point is changed. And determine the first performance in a second predetermined manner if the performance of the data point is less than the global sphere baseline standard indicator To change the quasi, change the radius length of the three-dimensional sphere associated with the data point according to a second predetermined rule, and (3) determine the radius for the data point and the performance of the data point In order to maintain the current state of the first performance criteria when does not change, the radius length of the three-dimensional sphere associated with the data point is maintained according to a second predetermined rule. The system and method of the present invention may further include selecting a data point according to the generation of data points from T ₀ to T _N.

  According to an alternative embodiment of the invention, a visual analysis method for determining the occurrence of a data set is provided. The method includes receiving input from one or more data feeds or databases, creating a virtual three-dimensional global sphere, and first for placing data points in one of the longitude segments. Segmenting a virtual three-dimensional global sphere according to a longitude portion to define a reference for a latitude and a latitude for defining a second reference for placing a data point in one of the latitude segments Segmenting the virtual three-dimensional global sphere according to the directional segment.

This alternative embodiment further includes determining the position of each data point relative to the virtual three-dimensional global sphere from information about each data point received by the computer from one or more data feeds or databases, and a time T _0. Representing each data point at a surface location for a virtual three-dimensional global sphere as a three-dimensional sphere having a radius determined by a predetermined rule and at least by a first performance criterion, and one or more data feeds or Tracking the occurrence of each data point from time T ₀ to time T _N according to information about each data point received from the database and displaying on the display one of the following for each data point: Can: (1) Data points In order to change the first performance criteria in a first predetermined manner when the performance of the data point exceeds the global sphere baseline standard indicator Changing the length of the radius of the associated 3D sphere, (2) determining a radius for the data point, and a second predetermined method if the performance of the data point is less than the global sphere baseline standard measure In order to change the first performance criterion, change the radius length of the three-dimensional sphere associated with the data point according to a predetermined rule, (3) determine the radius for the data point, and Associated with the data point according to a predetermined rule to maintain the current state of the first performance criterion when the performance of the Maintain the radius of the length dimension sphere. Alternative embodiments may further include selecting a data point according to the generation of data points from T ₀ to T _N.

  According to an alternative embodiment of the present invention, a baseline standard presented as a global sphere that is segmented according to a predetermined rule for separating a data set for placement in a segmented portion of the global sphere A visual analysis method for determining the occurrence of a dataset for is provided. This alternative embodiment includes the steps of receiving input from one or more data feeds or databases and the radius of the global sphere that serves as a baseline standard measure by which the occurrence of the data set will be measured. Creating a three-dimensional global sphere having a global sphere center in accordance with a first predetermined rule for determination and displaying it on a display electronically connected to a computer. Further, according to this alternative embodiment, the global sphere is segmented according to the longitude portion to define a first reference for placing a data point in one of the longitude segments, and the latitude direction Segmented according to a latitude segment to define a second criterion for placing a data point in one of the segments, the location of each data point by a computer from one or more data feeds or databases A global sphere is determined from information about each received data point.

The system and method of the present invention further provides each data in surface position for a global sphere as a three-dimensional sphere having a distance from the global sphere center determined by a second predetermined rule and at least by a first performance criterion. Tracking the occurrence of each data point from T ₀ to T _N according to the step of representing the point at time T ₀ and information about each data point received from one or more data feeds, of the following for each data point: And (1) the computer determines the distance from the global sphere center for the data point, and the performance of the data point exceeds the global sphere baseline standard indicator. To change the first performance criteria in the first predetermined way Changing the distance from the global sphere center of the three-dimensional sphere associated with the data point according to a second predetermined rule, (2) the computer determines the distance from the global sphere center for the data point, and the data A three-dimensional associated with the data point according to a second predetermined rule for changing the first performance criterion in a second predetermined manner when the performance of the point is less than the global sphere baseline standard indicator Changing the distance of the sphere from the global sphere center, (3) determining the radius for the data point, and maintaining the current state of the first performance criteria if the data point's performance remains unchanged Maintain the distance from the global sphere center of the 3D sphere associated with the data point according to a predetermined rule. This alternative embodiment may include selecting a data point according to the generation of data points from T ₀ to T _N.

  The system and method of the present invention provides a visual analysis embodiment for interactive analysis of multidimensional data stored in a database. This embodiment can include receiving multidimensional data from a database, wherein the multidimensional data is characterized by data values associated with at least a name, a type, and at least a temporal point, the multidimensional data being Have the same name associated with the same entity, and the multidimensional data has the same type associated with the same category.

  This embodiment may further include providing a system user interface on the display for display of interactive analysis of the multidimensional data, the system user interface displaying the virtual three-dimensional space and receiving the multi-dimensional data received from the database. A baseline sphere in a virtual three-dimensional space generated by a computer based on the dimensional data is displayed on a display, the radius of the baseline sphere from the center of the virtual baseline sphere corresponds to the first index, It can be segmented by longitude and latitude parts. According to this embodiment, the method may further include calculating a corresponding longitude direction and latitude direction coordinate value for each multidimensional data, and the longitude direction and latitude direction coordinate value of each multidimensional data Calculated based on the multidimensional data name that defines the coordinates and the associated multidimensional data type that defines the latitude coordinates, and for each multidimensional data, displays the sphere corresponding to each multidimensional data, each sphere is the longitude The radius of the sphere arranged in the virtual three-dimensional space according to the direction and latitude coordinate values and associated with each multidimensional data is calculated based on the data value for the multidimensional data, and the center of each multidimensional data sphere The distance from the center of the baseline sphere to the center of the baseline sphere is calculated based on the data values for the multidimensional data, and is calculated according to its position relative to the baseline sphere Selecting multidimensional data by scan of the device.

  The system and method of the present invention provides for the data set to a baseline standard presented as a global sphere segmented according to a predetermined rule for separating the data set for placement in a segmented portion of the global sphere. Included is an embodiment of a system for determining occurrence. The system can receive input from one or more data feeds or databases to determine the radius of the global sphere that serves as a baseline standard measure by which the occurrence of the data set will be measured. A 3D global sphere is created and displayed on the display according to a first predetermined rule, and the longitude portion is defined to define a first reference for placing data points in one of the longitude segments. Thus, the global sphere can be segmented, and the global sphere can be segmented according to the latitude segment to define a second criterion for placing a data point of one of the latitude segments.

This embodiment of the system further determines the position of each data point relative to the global sphere from information about each data point received by the computer from one or more data feeds, and determines the radius for the global sphere. Each data point in the surface position for the global sphere can be represented at time T ₀ as a three-dimensional sphere having a radius determined by two predetermined rules and at least by the first performance criterion.

Furthermore, this embodiment of the system tracks the occurrence of each data point from time T ₀ to time T _N according to information about each data point received from one or more data feeds, and for each data point , (1) changing the radius length of the three-dimensional sphere associated with the data point according to a second predetermined rule for determining the radius for the data point, so that the performance of the data point is a global sphere baseline standard Changing the first performance criterion in a first predetermined manner when the index is exceeded, and (2) determining the radius for the data point, and the performance of the data point is less than the global sphere baseline standard index In the second predetermined method in order to change the first performance criteria related to the data point according to the second predetermined rule Changing the length of the radius of the 3D sphere, and (3) determining the radius for the data point and maintaining the current state of the first performance criteria if the performance of the data point does not change To display on the display one of maintaining a radius length of the three-dimensional sphere associated with the data point according to a second predetermined rule, from T ₀ to T _N Data points can be selected according to the occurrence of the data points.

  According to an alternative embodiment, a visual analysis system for determining the occurrence of data is provided. The system can receive input from one or more data feeds to create a virtual 3D global sphere and to define a first criterion for placing data points in one of the longitude segments Segment the virtual 3D global sphere according to the longitude part and segment the virtual 3D global sphere according to the latitude segment to define a second criterion for placing data points in one of the latitude segments To do.

An alternative embodiment of this system further determines the position of each data point relative to the virtual three-dimensional global sphere from information about each data point received by the computer from one or more data feeds, at time T ₀ , Each data point is represented on a display at a surface position for a virtual 3D global sphere as a 3D sphere having a radius determined by a predetermined rule and at least by a first performance criterion, and received from one or more data feeds Track the occurrence of each data point from time T ₀ to time T _N according to the information about each data point, and for each data point, (1) a predetermined rule for determining the radius for the data point Therefore, the 3D sphere associated with the data point Changing the length of the diameter and changing the first performance criteria in a first predetermined manner when the performance of the data point exceeds the global sphere baseline standard index; and (2) the radius for the data point. And associated with the data point according to a predetermined rule to change the first performance criterion in a second predetermined manner when the performance of the data point is less than the global sphere baseline standard indicator Changing the radius length of the 3D sphere; and (3) determining the radius for the data point and maintaining the current state of the first performance criteria if the performance of the data point does not change, One of maintaining a radius length of the three-dimensional sphere associated with the data point according to a predetermined rule is displayed on the display, from T ₀ to T _N The data points can be selected according to the occurrence of the data points.

  According to yet another embodiment, a data set for a baseline standard presented as a global sphere segmented according to a predetermined rule for separating the data set for placement in a segmented portion of the global sphere A visual analysis system is provided for determining the occurrence of. An alternative embodiment of this system receives a radius from a global sphere that serves as a baseline standard indicator that receives input from one or more data feeds or databases, thereby measuring the occurrence of the data set. A three-dimensional global sphere having a global sphere center is created according to a first predetermined rule for determination and displayed on a display electronically connected to a computer, with a data point in one of the longitude segments. Segment a global sphere according to a longitude portion to define a first reference for placement and latitude to define a second reference for placing a data point in one of the latitude segments The global sphere can be segmented according to the direction segment.

An alternative embodiment of the system further determines the position of each data point relative to the global sphere from information about each data point received by the computer from one or more data feeds, and at time T ₀ , a second Each data point can be represented in a surface position for a global sphere as a three-dimensional sphere having a distance from the global sphere center determined by a predetermined rule and at least by a first performance criterion.

Furthermore, this alternative embodiment tracks the occurrence of each data point from time T ₀ to time T _N according to information about each data point received from one or more data feeds, and for each data point (1) determining the distance from the global sphere center for the data point, and changing the first performance criterion in a first predetermined manner when the performance of the data point exceeds the global sphere baseline standard index Changing the distance from the global sphere center of the three-dimensional sphere associated with the data point according to a second predetermined rule; and (2) determining the distance from the global sphere center for the data point. , If the performance of the data point is less than the global sphere baseline standard indicator, To change, the distance from the global sphere center of the three-dimensional sphere associated with the data point according to a second predetermined rule, and (3) determining a radius for the data point, the data point Maintaining the distance from the global sphere center of the three-dimensional sphere associated with the data point according to a second predetermined rule to maintain the current state of the first performance criterion when the performance of the Can be displayed on the display, and data points can be selected according to the occurrence of data points from T ₀ to T _N.

  According to another embodiment of the system and method of the present invention, a visual analysis system for interactive analysis of multidimensional data stored in a database is provided. The system can receive multidimensional data from a database, where the multidimensional data is characterized by data values associated with at least a name, a type and at least a temporal point, and the multidimensional data is associated with the same entity. The multi-dimensional data has the same type associated with the same category and provides a user interface for displaying interactive analysis of the multi-dimensional data, the user interface is a virtual three-dimensional Display a space, display a baseline sphere in a virtual three-dimensional space generated by a computer based on multidimensional data received from a database, and the radius of the baseline sphere from the center of the virtual baseline sphere is a first indicator The baseline sphere is It can be segmented.

  According to this alternative embodiment, the visual analysis system can calculate the corresponding longitude direction and latitude direction coordinate values for each multidimensional data, and the longitude direction and latitude direction coordinate values of each multidimensional data are Calculated based on the multidimensional data name that defines longitude direction coordinates and the associated multidimensional data type that defines latitude direction coordinates, and for each multidimensional data, a sphere corresponding to each multidimensional data can be displayed , Each sphere is arranged in a virtual three-dimensional space according to the longitude and latitude coordinate values, and the radius of the sphere associated with each multidimensional data is calculated based on the data value for the multidimensional data, The distance from the center of the data sphere to the center of the baseline sphere is calculated based on the data values for the multidimensional data and made its position relative to the baseline sphere It can be selected multidimensional data by a computer-based system I.

  For these and other embodiments, the invention will be described in more detail hereinafter with reference to the drawings.

FIG. 1 shows a representative visualization of large data according to the prior art method. FIG. 2 shows a representative visualization of the entire data set with each data point presented in a single region represented by a sphere according to an embodiment of the present invention. FIG. 3 illustrates an exemplary visual analysis process according to an embodiment of the present invention. FIG. 4A shows a representative visualization of mutual fund data according to an embodiment of the present invention. FIG. 4B shows a representative visualization of mutual fund data according to an embodiment of the present invention. FIG. 4C shows a representative visualization of mutual fund data according to an embodiment of the present invention. FIG. 4D shows a representative visualization of mutual fund data according to an embodiment of the present invention. FIG. 4E shows a representative visualization of mutual fund data according to an embodiment of the present invention. FIG. 4F shows a representative visualization of mutual fund data according to an embodiment of the present invention. FIG. 4G shows a representative visualization of mutual fund data according to an embodiment of the present invention. FIG. 4H shows a representative visualization of mutual fund data according to an embodiment of the present invention. FIG. 5A shows a representative visualization of a regional workspace according to an embodiment of the present invention. FIG. 5B shows a representative visualization of a regional workspace according to an embodiment of the present invention. FIG. 5C shows a representative visualization of a regional workspace according to an embodiment of the present invention. FIG. 5D shows a representative visualization of a regional workspace according to an embodiment of the present invention. FIG. 5E shows an exemplary visualization of the timeline aspect of the present invention. FIG. 5F shows a representative visualization of the timeline aspect of the present invention. FIG. 5G shows an exemplary visualization of the timeline aspect of the present invention. FIG. 5H shows an exemplary visualization of the timeline aspect of the present invention. FIG. 6A shows a representative visualization of a survey space according to an embodiment of the present invention. FIG. 6B shows a representative visualization of the survey space according to an embodiment of the present invention. FIG. 7 shows an exemplary data model for visualizing the growth rate of mutual funds in the US market according to an embodiment of the present invention. FIG. 8 shows a representative visualization of an overview of the US mutual fund market in 2012 according to an embodiment of the present invention. FIG. 9 shows a representative visualization of a designated domestic institutional investor (“QDII”) in 2018 with a comparison diagram according to an embodiment of the present invention. FIG. 10 shows a representative visualization of all mutual funds shown in the area according to an embodiment of the present invention and QDII forecast results for China in 2018. FIG. 11 shows a representative visualization of the Chinese mutual fund market according to an embodiment of the present invention. FIG. 12A illustrates an exemplary external use example of the system and method of the present invention. FIG. 12B shows an example of an exemplary internal use of the system and method of the present invention. FIG. 13 shows an exemplary visualization of the operational risk data model. FIG. 14A shows a representative visualization of portfolio risk and performance evaluation according to an embodiment of the present invention. FIG. 14B shows a representative visualization of portfolio risk and performance evaluation according to an embodiment of the present invention. FIG. 14C shows a representative visualization of portfolio risk and performance evaluation according to an embodiment of the present invention. FIG. 14D shows a representative visualization of portfolio risk and performance evaluation according to an embodiment of the present invention. FIG. 14E shows an exemplary visualization of portfolio risk and performance evaluation according to an embodiment of the present invention. FIG. 14F shows a representative visualization of portfolio risk and performance evaluation according to an embodiment of the present invention. FIG. 14G shows a representative visualization of portfolio risk and performance evaluation according to an embodiment of the present invention. FIG. 14H shows a representative visualization of portfolio risk and performance evaluation according to an embodiment of the present invention.

  The present invention is directed to a visual analysis system and method for providing interactive analysis of multidimensional temporal data. The disclosed system and method utilize a novel 3D visualization that allows a compact and universal overview of large-scale data and the ability to deepen further details. The system can extend the capabilities of existing mapping techniques by visualizing domain data based on 3D geometry enhanced by size, color, motion and sound. The system is designed with customizable visualization that can be adapted to different data models and applied to multiple domains. The system adapts to different data models by using a set of virtual regions and spheres to visualize the data and utilizes longitude, latitude and distance from the center of the sphere to define the data model. The differential perspective view and the provided filter can be adapted to provide visualization of different data models.

  When the visualization system and method are used for the financial industry, for example, in the context of temporal and spatial relationships to benchmarks such as the S & P 500, portfolio risk by investment manager, industry sector, asset class, geography or other factors And a performance evaluation can be provided. The system and method of the present invention can also be used for visualization operational risks in an enterprise management environment.

  When the system and method of the present invention is first used for visualization of financial industry segments such as US mutual funds, it is a visualization of the entire US mutual fund industry in a single geometry based on the sphere domain. Have the ability to. Each sphere represents a fund data point and is enhanced by size, color, movement and sound. The system can use a time axis to visualize a static region in a fixed time frame, while simultaneously showing the dynamic occurrence of the region over any period of time.

  As an example, on December 31, 2011, the US mutual fund industry had approximately 21,000 fund stock classes with a combined total of 765 fund management companies and a total asset of 9.8 trillion. As of June 30, 2013, the US mutual fund industry had 23,000 fund stock classes and 11 trillion assets. The system and method of the present invention can present the development of mutual fund markets with dynamic 3D visualization as shown.

  Compared to prior art systems, the disclosed system is more flexible and can be applied to different areas where multi-dimensional and temporal data exist. The system and method of the present invention provides a new 3D visualization of large scale data in a universal overview, large scale data management, temporal data visualization and interactive analysis of data.

  The visual analysis system and method of the present invention can access a computer's graphics processing unit (GPU). For example, the system and method can access a library of GPU routines and functions to efficiently handle the visualization of large data sets. GPUs allow computers to render complex 3D computer animations.

Visualization Techniques Visual Encoding The system and method of the present invention visualizes large data in a vast area. The components of this area include global benchmarks and data points that are segmented and distributed with respect to global benchmarks according to predetermined rules as described.

  Referring to FIG. 2, generally at 200, an overview of the entire data set is presented in a single area. Each data point is represented by a sphere. Each sphere will have a different size in the 3D domain based on predetermined rules such as the net asset value (NAV) of mutual fund data points and can be enhanced by color, movement and sound. For example, in FIG. 2, a data point sphere having a first color outside the large global sphere is superior in performance to the operation of the large global sphere with such data points as standard indicators. Data point spheres with a second color that are inside large global spheres may be inferior to large global spheres where these data points serve as standard indicators Can be shown. The attributes of the multidimensional data can be mapped to these visual encodings by an optional data model, thus allowing a compact overview of the entire data set.

  The system and method of the present invention can use a time axis to visualize temporal data. In addition to displaying static regions within a fixed time frame, the system can display the dynamic occurrence of regions over any period of time.

Customizable Visualization The system and method of the present invention can be customized by providing data visualization. Reparameterization can result in completely different data models and visual representations even using the same data and visualization techniques. By changing the visual parameter settings, the user can obtain different visual impressions and have many opportunities to gain insight into data patterns, trends and correlations. Customizable visualization is achieved by designing different data models for different user requirements and use cases. The system and method of the present invention allows a system user to set up his workspace with a selected data set and build a data model using customizable parameters.

Interaction Design The system and method of the present invention allows a user of the system to have multiple interactions with data visualization. These interactions can provide valuable information and support analysis of large data sets. A better understanding of the data set is built, tested, refined, and shared by interactive manipulation of the visual interface. A significant amount of information regarding the analysis process of the system user by the visualization tool is captured by such interactions. The system and method of the present invention implements some of the interactive features in the following manner.

  Data filter: By default, the region gives an overview of the entire data set. The system user can select a particular type of data that is of interest, and other data is filtered from the region.

  Navigation: Since the spheres are distributed in a 3D region and there is an earth frame, the spheres behind or inside the earth frame can be hidden by things on the front or outside of the earth. The system and method of the present invention allows a system user to rotate the earth to explore it from different angles. System users can also zoom in by having a more detailed view of some special groups of spheres or spheres inside the earth.

  On-demand details: When a sphere is far away from the center of the region, it can have special attribute values that are very different from others. These “outliers” can have great potential advantages for further investigation. The system user can click on the sphere and drill down for detailed information displayed in a small window. Furthermore, an HTTP link associated with the sphere can guide the user to a website that provides information about the sphere and other external resources.

  Auto play: There is a time axis at the bottom of the area. As the time axis is dragged, the region changes accordingly and presents the state of the data set corresponding to the particular time selected using the time axis. However, it may be difficult to present a continuous change over a long period of time by manual operation. Thus, the system and method of the present invention provides an autoplay feature that automatically moves along the time axis with an optional speedup, and therefore the system user focuses on the analysis of patterns, trends and correlations. It is possible to do.

  Referring to FIG. 3, generally at 300, the visual analysis method of the present invention is shown. The visual analysis process combines automatic analysis and visualization methods by tight coupling via system user interaction to obtain knowledge and information from the data. FIG. 3 shows an abstract overview of the stages and transitions in the visual analysis process.

  In many application scenarios, raw data 301 from disparate data sources needs to be integrated before visual or automated analysis methods can be applied. Raw data often contains different types of errors, and some of the data is incomplete. Thus, the first step at 302 is to preprocess and transform the raw data to derive different representations for further investigation. During the process, pre-analysis and validation is used to ensure the quality of the data. Other typical preprocessing tasks include data reformatting, data interpolation, data cleaning, normalization, grouping and integration.

  In step 303, data abstraction can be used to generate an abstract data set independent of the corresponding source. After transformation, data characterization (step 304) needs to be clearly defined so that important attributes can be extracted for visualization and analysis. After characterization at 304, the data analyst can select the visual or automated analysis method or both to apply. When automatic analysis is first used, data mining methods are applied to build a model of the data. Once the preliminary model is created, the data analyst can evaluate and review the model that can be done by interacting with the data. Visualization allows data analysts to interact with automated methods by modifying parameters or selecting other analysis algorithms.

  The model visualization 305 can be used to evaluate the generated model 306 survey results. This process results in continuous improvement and validation of the preliminary model.

  Using the present invention, misunderstandings in intermediate steps can be detected early, resulting in a better data model and higher confidence. When a visual data search is first performed, the system user must confirm the hypotheses generated by the interactive analysis. System user interaction with visualization can reveal insightful information, for example, by zooming in on different data regions or considering different visual views on the data. Views in the visualization can be used to steer model building when automatic analysis is performed. Thus, in the visual analysis process, knowledge 307 can be obtained from visualization, automated analysis, and previous interactions between visualization, models and data analysts.

Representative Research Representative research using the system and method of the present invention has been conducted using real world data within the financial domain of the Chinese mutual fund industry from analysis performed by the analysis group.

  The analysis group predicted that by late 2015, China's managed financial services industry total assets (“AUM”) (including public and private funds) would approach US $ 1 trillion. Such high asset volumes may not be permanently sustainable, but provide a basis for additional investment in the industry in some of the managers. This is supported by growth in the private sector, which is itself driven by high net worth and increased demand from institutional investors. A portfolio that is very unbalanced, both geographically and within asset classes, motivates this shift. QDII-a program where domestic investors buy global securities-also has room for growth given its small current size. Global investors are also seriously underweighted against China, with portfolio rebalancing and the emergence of Greater China's asset class (apart from BRICS (Brazil, Russia, India, China and South Africa)) Provides an important driving force for such transitions.

  There are currently several important and large pools of assets in mainland China that have not been further worked on. These range from insurer capital to pensions and other concentrated funds that are now very concentrated in cash and bonds. These imbalances alone do not mean that investments will be reallocated. Rather, the gap serves as the basis for assumptions on how asset flows are redirected over the next few years to seize potential opportunities for Chinese financial services industry entrants or entrants. In addition to being directly involved in the financial services industry, growth has also become more important as it is currently largely underdeveloped within the domestic fund industry and companies are facing rising costs. Provide support for supplementary and tertiary services.

  However, there is still a lack of powerful ways to explore and demonstrate this entire process and to support decision making and policy making. Traditional analysis with “basic” visualization methods is based on a compact overview of the entire mutual fund market and the lack of a “sophisticated” visualization method of detailed information about a single or specific group of mutual funds. Can not provide other analysis tools driven by. Visual analysis, combining both highly effective visualization and interaction facilities and automatic intelligent data analysis methods, can transform large, complex data into reliable and understandable knowledge.

  The system and method of the present invention can visualize the entire mutual fund market while retaining as specific a single mutual fund information as possible. The disclosed system and method can also visualize dynamic changes in the market over time. The disclosed systems and methods can also support the work of domain analysts.

Data Characterization The first step in designing a visual analysis system according to the present invention is to define the purpose and characterization of the data that determines the visual encoding selection and data model. Using mutual funds as an example, mutual funds are a type of professionally managed collective investment vehicle that pools money from many investors to purchase securities. There is a wide range of fluctuations in mutual fund data that is always high dimension and changes dynamically. The characteristics of an investment trust for the purposes of the system and method of the present invention are as follows.

  Net asset value (NAV): Because the shares of such funds registered with the US Securities and Exchange Commission are redeemed at their net asset value, the assets of mutual funds are often associated with open ends or mutual funds. Is the value of debt minus the value of debt.

  Growth rate and volatility: These measure the performance of mutual funds. Negative growth rates are usually considered “bad”. Investment trusts with high volatility often mean high risk and high return, making it difficult to determine.

  Fund type: Since mutual funds are a broad concept, all mutual fund data is classified into seven types according to their main investments. They are stock funds, blend funds, qualified domestic institutional investors (QDII), fixed income funds, short-term debt funds, money market funds, closed-end funds. QDII is a method related to capital markets that is set up to allow financial institutions to invest in offshore markets such as securities and bonds. For illustrative purposes, QDII refers primarily to China QDII herein.

  Client: This is a financial company that runs a specific mutual fund.

  The main analysis of actual data can be done before visualization. Analysis of all Chinese mutual funds' NAVs reveals very large gaps between different mutual funds. The NAV ratio can be approximately 1000: 1 when comparing the maximum and minimum investment trusts. This is a very important characteristic of the Chinese mutual fund market that should be considered when building the data model.

Data Management Data management is important for VA tools and aims to ensure data consistency, avoids duplication and processes data transactions. Most of the data in the disclosed system is obtained from different websites where the format and content of the data can vary greatly. These heterogeneous data need to be preprocessed, integrated and validated before applying visualization techniques to build the data model.

  For illustrative purposes only, data was obtained from two data sources: (i) CSRC for Chinese fund data and (ii) Morningstar for US fund data. Apache's httpclient, an open-source scrolling tool, was used to acquire the data. The data can include, for example, the fund name, fund type, fund code, establishment date, and fund value for example for the entire 2012 year. The use of CSRC and Morningstar is for a specific use of visualization systems and methods related to the fund industry. It can be appreciated that other databases can be used as data sources. For example, use of the disclosed systems and methods is not limited to the financial industry. Other industries can utilize the disclosed embodiments. For example, the disclosed systems and methods can provide useful data analysis in commercial, transportation, or any other situation to provide visualization of large data sets.

  Preferably, data quality issues are dealt with after data is collected because the data is incomplete and inconsistent or may contain measurement errors. A “weighted voting” strategy is adopted to address quality issues. Weighted voting represents a generalization of the basic quorum-based scheme. For example, there are four main publicly available data sources in the Chinese mutual fund market named CSRC, JRJ, HeXun and Yinghe. Each data source is assigned a weight according to its reliability. The other three sources are low and have similar weights, but the official data source CSRC should have a higher weight because it is more reliable. If there are different data values between these data sources, each source “votes” for its own version. The score for the version of the data is a weighted vote summary, and the version with the highest score is selected. In most cases, the data from the CSRC is reliable and rules are defined to identify potential problems. If no CSRC version is selected and the difference between the CSRC and the voted is significant, an alert is issued to the observer.

Basic Data Model Details regarding the selection of a specific visual encoding and how to build the basic data model as intuitive as possible are presented below.

Visualization of net asset value The volume of a sphere is a reasonable and intuitive way to represent the net asset value (NAV) of an investment trust. However, the actual data characteristics indicate that NAVs cannot be directly mapped to volumes, for example, due to scaling. Small spheres are obscured by some very large spheres, thus destroying the delicate sensation of the entire market. According to the present invention, a log function is assigned to each investment trust's NAV. Using the log function, small spheres and large spheres can be clearly distinguished from each other while the entire region looks more harmonious.

However, if they are distributed in 3D space, it is difficult to compare the volume of two spheres, both small or both large. The question is, how do you “distribute” these spheres and what does their location mean in the region? Can be included. As described below, the system and method of the present invention addresses these issues. First, the center point of the region is added. The distance from the sphere to the center of the region is based on the volume of the sphere (ie, the mutual fund NAV). The large sphere is far from the center of the area. However, if the NAV is mapped directly to distance, many small spheres stay close to the center of the region, while the largest sphere is so far away that it cannot be displayed on the screen (the maximum ratio is About 1000: 1). The large space between the large and small spheres is empty, making the area unacceptable. The system and method of the present invention use the following basic calculation formula for calculating the distance from the investment trust to the center of the region.

  Here, V is the net asset of the investment trust. D is the distance from the investment trust to the center of the area.

The system and method of the present invention calculates the radius of the baseline sphere by the following equation:

である。ベースライン球の半径及び投資信託から領域中心までの距離は、画素単位で表示領域に線形的にマッピングされる。例えば、ベースライン球の半径は、各画素が４を表す場合には３１０画素にマッピングされる。 Here, Avg (V) is the average net asset of all investment trusts. R is the radius of the baseline sphere. For example, if the sum of net assets of all investment trusts is $ 10,000B, and there are 20,000 investment trusts, Avg (V) is $ 500M and R has a value of 1240

It is. The radius of the baseline sphere and the distance from the mutual fund to the center of the area are linearly mapped to the display area in pixel units. For example, the radius of the baseline sphere is mapped to 310 pixels if each pixel represents 4.

  By using this equation to calculate the distance between the sphere and the center of the region, the ratio of maximum distance to minimum distance is reduced to about 9: 1. This transformation makes the overall distribution more rational for visual analysis purposes. Here, the spheres can be divided into different groups according to their distance to the region center. For spheres in the same group, they have the same distance to the center of the region and are distributed on the surface of the same sphere that can be considered a virtual sphere.

  The radius, and thus the size, of each sphere representing the mutual fund can be customized by the system user and can be selected, for example, to provide a good visualization result that does not hide other spheres. Under an alternative embodiment, the radius of each sphere can be mapped to a specific business significance parameter or another dimension of data. Different algorithms can be used to create a mapping between the sphere / star radius and specific business significance parameters (data dimensions). For example, linear mapping can be used. Alternatively, log linear mapping can be used if linear mapping does not provide an acceptable visualization effect.

Fund type and client visual mapping Since other properties of mutual fund data need to be visualized, spheres cannot be randomly distributed on the surface. System users are familiar with the longitude and latitude directions when talking about the sphere. The system and method of the present invention uses this understanding in formulating a new visualization encoding. As described above, all mutual fund data are classified into seven types according to their main investments. Each type of mutual fund is mapped at different latitude intervals depending on the market share. Longitude represents the client of each mutual fund. A new sphere is added to the region to give the user an intuitive sense of “virtual index”. The center is the center of the region, and the volume of this special indicator sphere is also significant. It represents the average NAV of the entire mutual fund market so that the radius of this indicator has the same aspect ratio as the sphere. The indicator can be a “benchmark” and the extraterrestrial sphere has a higher NAV than the market average. Therefore, a system user can have a general idea of NAV size by comparing the NAVs of different investment trusts by checking the distance to the center of the region and checking the positional relationship between the index and the center. The distance from the sphere to the center of the region and the volume of the sphere both represent the NAV of the mutual fund. This visualization method is referred to as “redundant coding”, which can help improve accuracy and improve recognition.

  FIG. 4A shows an exemplary visualization of a large dataset according to an embodiment of the invention, generally at 400A. As described above, the system and method of the present invention uses a sphere to present a metadata set, and latitude and longitude dimensions are used to identify points in space. The circle passing through the two poles is a longitude line. A circle parallel to the equator is a line of latitude. All data points are presented on a large sphere (Earth). For mutual fund data sets, each sphere represents one fund. The sphere radius, sphere color, and sphere size can all be used to represent mutual fund performance data. Under an alternative embodiment, the system and method allow visualization of sound-based occurrences. For example, the system can use an uptone to specify that the value is increasing, while a downtone can specify that the value is increasing.

  In FIG. 4A, funds belonging to the same fund management company are arranged in the same longitude direction portion, and funds belonging to the same industry classification are arranged in the same latitude direction portion. The system and method of the present invention uses this new mapping algorithm to present an accurate view of the entire mutual fund industry and intuitive navigation tools for rotating, zooming in and out on the domain of data. use.

  Referring to FIGS. 4B and 4C, the S & P 500 indicator is used to represent the earth (baseline globe). As shown in FIG. 4B, funds belonging to the same fund management company are arranged in the same longitude direction portion, and as shown in FIG. 4C, funds belonging to the same industry classification are arranged in the same latitude direction portion. The For example, “red” can be used to indicate funds that are performing poorly, and “green” can be used to indicate funds that are above the indicator (baseline sphere). Funds with poor performance are placed within the global sphere, while funds with good performance are placed outside the global sphere. The system and method of the present invention provides an intuitive navigation tool for rotating, zooming in and out on an area of data.

  Referring back to FIG. 4B, the system user can filter the visualization based on a particular client. For example, FIG. 4B shows a fund from “State Street Global Advisors” at 402B. Also, what is shown in FIG. 4B is information about a fund having a fund ID XLY.

  Referring back to FIG. 4C, the system user can select a specific category of funds and display the corresponding funds in the area. For example, at 402C in FIG. 4C, it shows funds from two categories: “sector securities” and “taxable bonds”. As shown, at 402C, the system user has selected a specific fund with fund ID “FOCIX” belonging to the taxable bond category. When reviewing the visualization of these categories of funds, system users found that funds in the “Taxable Bonds” category are overwhelmingly inferior in the market because they are shown in red and in the earth. Can be clearly observed. On the other hand, funds in the “sector equity” category show more balanced behavior, and the number of funds in the balanced category is superior in performance and poor in performance in the market.

  With reference to FIGS. 4D-H, additional filtering interactions are shown. FIG. 4D shows the default workspace initialized with a complete region of data. FIG. 4E shows a longitude filter by selecting the names of several fund management companies. FIG. 4F shows a latitudinal filter by selecting the names of several categories. FIG. 4G shows the range filter in FIG. 4E using a range control user interface. FIG. 4H shows a complete cross-filtered view using longitude, latitude and range.

  Multiple workspaces allow users to name and identify independent study data sets. For example, the visualization in FIGS. 4D-H corresponds to a system user-defined workspace corresponding to the US mutual fund industry. As mentioned above, the ability to filter large data sets along longitude, latitude, and range (distance from the center) is an aspect of the interactive visual analysis system and method of the present invention. The work space can be created using filtering information from selection of longitude data (fund company), latitude data (category) or range. The ability to filter latitude, longitude and range information is referred to as a “cross-filtering” view.

  Referring to FIG. 5A, an exemplary visualization of a data set is shown, generally at 500A, and an area workspace 500 is selected for display. The area workspace has six visuals for analyzing the mutual fund industry: control bar 501A, area control panel 502A, data set area 503A, timeline scroll bar 504A, category panel 505A and selected fund data panel 506A. Shown to contain elements.

  5A-B present the entire US fund industry in a single geometry based on the area of the sphere. As described above, each sphere represents a fund data point and is enhanced by size, color, movement and sound. The current sample area shown in FIG. 5A begins on December 31, 2011. At that time, the US mutual fund industry had ~ 765 fund management companies with 9.8 trillion total assets and ~ 21,000 fund stock classes. At that point, all spheres are displayed on the surface of the earth. The sample area ends on June 30, 2013 with more than 23,000 fund stock classes and 11 trillion assets and is shown in FIG. 5B. Different colors can exhibit different mutual performance. For example, in FIG. 5B, funds above the baseline can be displayed in a first color, such as green, and are shown outside the baseline sphere. Funds below the baseline can be displayed in different colors, for example red, and are shown inside the baseline sphere.

  Referring to FIGS. 5C-D, the entire Chinese fund industry is visualized in a single geometry based on the area of the sphere in two different time periods. In FIG. 5C, all of the mutual funds are placed on the surface of the global sphere. FIG. 5D shows the distribution of those funds inside and outside the global sphere based on the performance in the snapshot at a later time.

FIG. 5E shows an exemplary visualization of the entire US fund industry, generally at 500E, where the duration of the analysis is set to be year to day (YTD) 501E. Therefore, on January 1, 2013, the start time T ^{0 of the} area is set. In the start time T ^0, as described above, all the funds are displayed on the surface of the base line sphere, for example, having the same color, such as yellow 502E. In FIG. 5F, generally at 500F, a system user using search field 501F of search tab 502F can search for funds that include the term “fact”. In response to the search, the system only filters all non-response funds and displays the funds identified in the search. In the visualization of FIG. 5F, there are 18 funds returned as shown in the fund name results area 503F and there is one fund management company (FMC) returned as shown in the FMC results pane at 516F. These funds are shown in area 504F.

The system user can also select a specific fund from the 18 funds. For example, the user can select a fund with fund ID “FSE”, and the sphere corresponding to the fund can be visualized in a different color, for example blue, corresponding to the fund that was not selected. It can be displayed as an enlarged sphere compared to the sphere. The details of the selected fund can also be displayed in window 506F. For example, window 506F may provide fund ID, fund name, fund category, fund management company, fund establishment date, fund NAV per share, fund movement and related fact sheets. The status of an index, for example the S & P 500 index, is shown in area 507F. Field 507F shows the indicator name 508F, the indicator percentile change 509F from the selected start time, the indicator value 510F, the overperformance fund percentile and absolute number 511F, and the underperformance fund percentile and absolute number. . As expected, the visualization of FIG. 5F corresponds to the start time T ⁰ , so the indicator percentile change 508F is 0%. In FIG. 5F, the radius of the baseline sphere 513F corresponds to the index number 510F at a specific point in the timeline. Timeline 514F points out the corresponding time. As shown in the timeline, the time is set to January 1, 2013.

Figure 5G, in general 500G, shows the change in the data set to a time _{T 1.} The selected fund's performance 505F is superior to the baseline and is located outside the baseline surface 513F. FIG. 5G also shows that the index has increased by 8.76% (509F) to a value of 155.18 (510F). The increase in the index is also indicated by the larger radius of the baseline sphere 513F compared to the radius of the baseline sphere 513F in FIG. 5F. FIG. 5G also shows that 11 or 61.1% of the 18 selected funds are above the baseline (511F) while the remaining 7 are below the baseline (512F). Eleven funds above are displayed in green and seven below are displayed in red. Field 506F provides current information about the selected fund. For example, the NAV per share has been updated to a value of 12.4 to correspond to the current time, which is March 8, 2013, as shown in the timeline 514F. The amount of movement is updated to a value of 22.4%. FIG. 5G also shows fund 515F displayed in red that is clearly below the baseline index.

FIG. 5H shows the continuous change of the selected fund at time T ₂ and the performance of the selected fund, generally at 500H. Region 506F provides current information about the selected funds in time T _2. For example, the NAV per share is updated to a value of 12.45 so as to correspond to the value of May 19, 2013 as shown in the timeline 514F. The moment value is updated to a value of 22.9%. Fund 515F that was displayed in below and red at time T _1, the current is shown above the baseline and is therefore displayed in green.

  Referring to FIG. 6A, generally at 600A, an exemplary visualization of a data set is shown and a search workspace is selected for display. A particular fund 601A can be selected and visualized in the workspace in a different color than other funds, such as blue.

  Referring to FIG. 6B, generally at 650, another exemplary visualization of a data set is shown, and a search workspace 650 is selected for display. As can be seen from FIG. 6B, if the system user selects the “Search” tab 651, a search dialog 652 is displayed and the system user can begin entering the name of a particular fund in the search field 653. In a particular example, the user has entered “s & p”. Search results are displayed in the search dialog 652. The system can match any metadata provided by the fund name or data vendor with the string entered in the search field.

Growth Rate and Volatility Visualization According to the present invention, the growth rate and volatility of mutual funds can be visualized in the domain. Since mutual fund data is time-series data, both growth and volatility are highly time related. A time axis is added at the bottom of the region to visualize changes in the mutual fund market over time. For a single sphere, when the user drags the time axis, it moves in response to changes in its NAV. For example, in 2011, the fund has a higher NAV than the average value of the entire market and stays outside the virtual earth. As described above, the virtual earth represents any index that can be selected by the system user. In 2012, if the fund's NAV shrinks below the average, it moves inside the virtual earth. Thus, the system user can have a general idea of growth rate and volatility by checking the moving speed and range of the sphere. Traditionally, mutual funds are often compared to stock market indicators such as the “Dow Jones Industrial Average” in the US and the “Shanghai Composite Index” in China. The growth rate of mutual funds is indicated by the color of the sphere. For example, in the disclosed method and system, when visualizing Chinese mutual fund data, red indicates a sphere above the market, while green indicates a sphere below the market. However, good and poor performance can be expressed in many different ways and still remain within the scope of the present invention.

  This exemplary data model aims at the common characteristics of a single mutual fund data that can also be visualized by traditional 2D graphs. However, the system and method of the present invention can present all mutual fund data on the same screen (ie, 3D domain), provide a compact overview of the entire mutual fund market and be as specific as possible. Holds information for a single mutual fund.

Customizable visualization The data model described above is primarily focused on fund NAV visualization. However, it does not provide information on fund growth rates compared to a specific date in the past. Referring to FIG. 7, generally at 700, the particular data model shown is designed to represent mutual fund growth rates. In this model, the distance from the sphere to the center of the region can be calculated by the following equation.

  R is the distance, and c is the growth rate of the investment trust compared to a specific date in the past. The growth rate can be negative, but since negative distances cannot be visualized in the region, all growth rate values are incremented by 1 to ensure that the value is always non-negative. Is done. It is then multiplied by a constant “α” (default value is 200) to increase the distance so that the region looks more reasonable. In this model, the “benchmark” (ie, virtual earth) no longer represents the market NAV. Here, it represents the growth rate of the stock market index (such as the “Dow Jones Industrial Average” in the United States) compared to a specific date in the past. The new formula for calculating the radius of the earth is the same as the mutual fund sphere. In this way, the user can easily compare the performance of investment trusts by the stock market index by confirming the position of the sphere. Extraterrestrial spheres outperform the market, while extraterrestrial spheres outperform the market. Green and red are also used to enhance cognition as described above.

  As described above, the data model of FIGS. 5A-B visualizes the growth rate of mutual funds in the US market. When the indicator is growing, most spheres with low latitudes are red, meaning they are below the market. Interactive analysis according to the system and method of the present invention shows that the type of “debt fund” account for the majority of red spheres indicates that the debt fund is less affected by the stock market.

Application Examples The following discussion shows two representative examples with actual data from both the Chinese and US mutual fund markets. The first example shows changes in the Chinese mutual fund market in 2012 due to interactive operations. The second example demonstrates the ability of the system and method of the present invention to predict the development of China QDII.

Visualization of mutual fund market The discussion above described how mutual fund data is obtained from different data sources. FIG. 8 shows an overview of the US investment trust market in 2012. The lower right corner shows the 2012 region of the Chinese market for comparison. After pre-processing and integration, these actual mutual fund data are applied to the system. There are 21,686 funds from the US and 869 funds from China. The United States has much more funds than China, but if the radius represents the average NAV, the difference in the illustrated earth radius is small because it is adjusted by the above formula. However, as can be seen from FIG. 8, the richer US market is still intuitive. The virtual earth corresponding to the US market is almost completely covered by a sphere. In the Chinese market, a part of the surface of the virtual earth, for example, the space for QDII is almost empty.

  Interactive operations are also fully considered in this case. By dragging the time axis or clicking the auto play button, the user can see changes in the mutual fund market. Over time, the spheres change their color and move in and out of the earth. If a particular mutual fund leaves the market, some spheres disappear and new spheres also arrive in the territory. With data filters, it is clear that QDII in China is still “stopping fire”. Some QDII “outliers” have also been found to have a much higher NAV than other QDII spheres. Further investigation of the company's background confirms potential business benefits and investment opportunities.

China QDII Forecast Analysis In early 2012, the total value of QDII in China was 9,070 million yuan, accounting for only 2.5% of the total investment trust market assets (340,740 million). China QDII is supposed to have a great potential advantage for custodian business, but simply “stop the fire”. The disclosed system and method allows for predictions on the mutual fund market. The goal is to show possible trends in the mutual fund market. In this case, the prediction model is based on the research activities of the Z-Ben advisor, the contents of which are incorporated herein by reference.

  According to the forecast model, the total NAV of China QDII grew to 53,690 million in 2018, about 6 times larger than 2012. Total market assets have grown to 1,55,250 million, much slower than QDII's growth rate. Since China QDII's capital has soared, dragging the time axis to a future time will create many new spheres of QDII in the region, as shown in FIGS. Autoplaying market changes from 2012 to 2018 reveals future trends for QDII. This visualization helps financial analysts make better decisions and persuade executives to adopt new policies.

  Specifically, FIG. 9 shows China's QDII in 2018 by a comparative view. All other types of funds are excluded from visualization. The lower right corner shows the region in 2012. It is clear from the visualization that China QDII is growing rapidly.

  Figure 10 shows the forecast results for China's QDII in 2018, with all mutual funds shown in the territory. The sphere representing QDII is highlighted in blue. Visualization reveals that both total NAV and QDII quantities have great potential for growth.

  Referring now to FIG. 11, generally at 1100, visualization corresponds to the Chinese mutual fund industry. As mentioned above, funds that exceed the index are shown in the first color. The under fund is shown in the second color, while the performance color according to the indicator is shown in the third color. As shown in FIG. 11, the user can move the mouse pointer over a specific fund type, eg, “stock fund”. This means that funds belonging to a specific fund type are displayed in the fourth color. Each representation of fund data points in FIG. 11 uses a “halo” effect rather than indicating the size of the fund based on a hard sphere.

  FIG. 12A shows external use cases and visualization functions for four different client segments. For example, the disclosed system can be targeted to C-suite executives and can provide useful visualization of asset allocation, performance factors and corporate risk management, and regulatory-related operations. The visualization function can be targeted for portfolio risk and performance assessment by managers or factors such as asset class, geography, sector, style and liquidity. Portfolio managers and retail investment advisors can be interested in asset selection and visualization of performance factors. The visualization function can also be targeted for portfolio risk and performance assessment by security or factors such as asset class, geography, sector, style and liquidity. Traders can be interested in asset allocation and transaction cost analysis. The visualization function can target security choices due to factors such as style and fluidity.

  FIG. 12B shows an internal use case and visualization function for three different client segments. Similar to FIG. 12A, the disclosed system can be targeted to C-suite executives and can provide useful visualization of asset allocation, performance factors, risk management and manager selection. The visualization function can be targeted for portfolio risk and performance assessment by managers or factors such as asset class, geography, sector, style and liquidity. Traders and electronic exchange businesses can be interested in asset allocation and transaction cost analysis. The visualization function can target security choices by factors such as style and fluidity. The general manager's strategy team can be interested in performance factors and risk assessment. The visualization function can target investor behavior by factors such as asset class, geography, sector, style and liquidity.

  The visualization system and method of the present invention can be used to identify the root cause of a problem and uses future time to restore the mean time between failure, service and many other attributes. Can be predicted. For example, a user can identify a particular group of systems that need the most attention. The disclosed system helps the user visualize the future risks of different groups of systems.

  Exemplary data models are sanity checks and risk predictions. The health check data model is used to indicate the failure rate, which is the time to restore the service of the incident that has already occurred and the time to pay for the service. The risk prediction data model can be used to predict future behavior based on events that have already occurred.

  All incidents come from different departments. Each department is mapped to different longitude intervals according to the proportion of all departments. Each incident category is mapped to a different latitude interval according to the proportion of all incident categories.

  Many issues related to the incident are the most important factors in operational risk management. The more problems of a particular type of incident, the more attention is required to solve it. If the incident type has many problems, it will have the shortest average time of the most likely failure interval, unless the average time to restore service is long.

  The disclosed system and method provide a great visualization of a number of problems. Since the same type of incidents are grouped together in a sphere, the user can visually identify which incident groups have many problems and which have a few problems. . For example, if one type of incident has more problems than any other type, certain portions of the area are filled with spheres, and the user can easily identify the incident.

  The recovery time is the time taken to recover after a failure occurs in the system. Incidents with long recovery times are more noticeable than incidents with short recovery times. Also, the recovery time can be interpreted as the age of the system. If the system has a long recovery time, it is likely that the system has been used for a long time. The size of the sphere can be used to represent the duration of the recovery time. Since the recovery time has a large range, the log function can be used so that the user can still see small spheres and the size difference between each sphere can still be apparent.

  Distance from the center of the region is the most intuitive way to represent the number of problems for a particular type of incident. It can be used to represent the reciprocal of failure rate or mean time between failures (MTBF) or the reciprocal of logarithmic MTBF. The radius of the global sphere is the average failure rate of all incidents. This helps the user to easily visualize the incident categories and departments with high failure rates and be most notable.

  The “burden on service capability” relates to the date and time when the incident occurred. Incidents that occur near the end of working hours will form more losses than incidents that occur during non-working hours. Color can be used to visualize the burden on service capabilities. For example, red can visualize a serious incident. Cases with a moderate burden turn yellow, and cases with a low burden turn green.

  For risk prediction data models, relevant attributes can include reliability, availability and maintainability. Reliability indicates how reliable the system is and can represent the probability that the system will survive in a given period of time. Availability indicates the probability that the system is operational when it needs to be used. Serviceability indicates how fast the system can be repaired after a failure and is the same as the average time to restore service.

The different mappings are summarized in the following table.

  An exemplary visual representation for an operational risk prediction data model can be shown generally at 1300 in FIG. FIG. 13 displays a sphere representing a data point and displayed in a three-dimensional space according to the department and case category of the corresponding data point. The mapping can be performed based on the above table. For example, the data points can be visualized as a yellow sphere 1301, a red sphere 1302, a blue sphere 1303 and a green sphere 1304.

  The visualization system and method of the present invention provides portfolio risk and performance by investment manager, financial industry sector, asset class, geography or other factors in the context of temporal and spatial relationships to benchmarks such as S & P 500, Shanghai Index, etc. Enables the construction and evaluation of The visualization system and method of the present invention also enables the construction and visualization of funds of funds (FoF).

  In general, FoF is an investment strategy that retains a portfolio of other investment funds rather than direct investments in stocks, bonds or other securities. Thus, the FoF portfolio construction and visualization according to the present invention relates to investments within funds that work together to form investment solutions. FoF portfolio construction involves combining fund investment strategies to address investment objectives and factors such as suitability for investment risk and expected life of the investment.

  There are two main considerations in building a FoF portfolio: (1) asset allocation related to how investments are spread across different asset types and regions, and (2) selected asset classes and Fund selection related to fund manager and fund selection to represent each of the sectors. In general, in the medium to long term, it is understood that asset allocation typically has a significant impact on fluctuations in the return of the FoF portfolio.

  In accordance with the system and method of the present invention, system users are provided with a method for visually analyzing FoF asset allocation over the medium to long term. This is described in more detail with reference to exemplary FIGS. 14A-H.

  In general, FIGS. 14A-H illustrate exemplary visualization of FoF asset construction, allocation, and change. Referring to FIG. 14A, generally at 1400A, the system user can form a FoF by pressing the “+” symbol 1404 on the FoF control panel 1402. Before a new FoF is formed, the initial state of the “Fund of Funds” control panel tab 1401 is initially empty as shown in FIG. 14A. Specifically, the FoF tab 1401 is used to indicate the current state of the name FoF selected from the name “Fund of Funds” control panel 1402. As shown in FIG. 14A, the selected analysis period is YOY (YoY) 1403. However, other time periods can be selected and are still within the scope of the present invention.

  FIG. 14B shows the FoF control panel 1402 after the system user has created the FoF, generally at 1400B. In the example shown in FIG. 14B, the system user created a FoF called “SnP baseline” in 1405. The control icon 1406 can be used to edit and delete the created FoF. For illustrative purposes only, the FoF created in FIG. 14B has an investment purpose to exceed the S & P 500 index using YOY analysis. Asset selection strategies can include analysis of S & P 500 funds across sector funds and listed mutual funds (ETFs).

  FIG. 14C shows a result 1406 from a query requesting all funds having the keyword “s & p”, generally at 1400C. FIG. 14D illustrates the selection of the first fund 1407 to add to a potentially new FoF, generally at 1400D, where the selected fund is the fund ID “UPRO” 1408. The system user has the option to open the FoF panel and add the selected fund to the SnP baseline FoF 1408 or another selected fund. FIG. 14E shows the addition of the first fund 1409 (UPRO) for the new “SnP Baseline” FOF, generally at 1400E.

  14F-H are generally selected at 1400F-H, respectively, for a particular asset type for a particular selected fund, selected and aligned along the system user's longitude portion and aligned along the latitude portion. 12 funds 1410 are shown. In addition, these 12 funds are shown in FIGS. 14F-H due to different advantages in different time periods. FIGS. 14F-H also show information for the selected fund “OBCHX” in different time periods.

  Investments in the name of the FoF portfolio are carried out according to the market. As the timeline progresses, the current asset allocation of the FoF portfolio drifts away from the original target asset allocation, for example, the level of impact of those favorable risks. If the portfolio is left unadjusted, the risk is too high or too conservative. Therefore, the FoF asset allocation is periodically rebalanced to return the current asset allocation in line with the original planned asset allocation target, for example, the level of favorable risk impact. The FoF function can provide a powerful visualization tool for building, monitoring and rebalancing asset allocations over selected time periods.

Multi-user collaboration The disclosed systems and methods allow real-time collaboration between multiple users in different roles, including real-time information sharing, content creation and distribution. A user of the system can implement a business process such as fund accounting through this collaboration function. System users with different roles can publish information, comments or status changes, for example, a fund accountant can publish fund processing status via the system front end. The system can propagate all changes to all backends in various different roles and to all frontend active users in real time based on their access control and subscription of interest. System users monitor or act on real-time updates of aspects of information from multiple sources via the system front end. Exemplary actions include approval, rejection, or task assignment. The system provides an integration option for external systems to allow users to interact with other external systems and participate in individual business processes.

Customizable visualization or perspective customization capabilities Users or operational staff can create new perspectives on the fly by creating mappings between visual parameters and business data elements, with long-term development periods Enable business-friendly views for users without intervention. The system may also allow the user to observe the same set of large data sets from various business significant perspectives, for example by yield, by total assets or by risk.

  The system can maintain a complete set of business data dictionaries and underlying mapped data sources. For each supported visualization model, visual parameters are specified for the user to understand the visualization model and subsequent mapping. The system user interface allows the system user to map business parameters to visual parameters, thus creating a perspective view of the business parameters.

  The system can, for example, expand a mapping, such as a linear mapping, a mapping to one or more enumerated values, for example, a numerical mapping such as a cube root, and a mapping of a business parameter to one or more visual parameters. Allows combinations. The user can easily switch perspectives during analysis of the selected data set and can view various aspects of the data set. In addition, the user can adjust the viewpoint mapping during the run to see the effect on the fly, eg by changing the sphere / distribution / color.

Real-time event function (event streaming)
The event animation function enables the capture and visualization of real-time external incidents such as, for example, market data, news, external systems or business events and significant internal changes such as meteor visual effects.

  The system has a back-end process, can calculate external incoming incidents, and can pre-calculate and capture important internal changes. The back end of the system can capture events to the front end user and the user can act on the event for each business process or make discretionary decisions.

  The system enables a generic meta-database data service infrastructure. A common meta-database data service infrastructure enables visual systems to visualize various potential sources of business data. The system can maintain a complete set of business data dictionaries and underlying mapped data sources. The system provides a service-oriented data service based on the data service descriptor provided to the data consumer. The data service descriptor may include, for example, a metadata description of necessary data, a data format (XML, JSON, etc.), a data provision channel (http, ftp), a data provision approach (ad hoc query, publication / subscription, etc.). .

Sound and Music Effect Function The system can include a value map for a specific musical sound at the corresponding frequency. The temporal data for each star object can be mapped to a single melody. The overall melody is a composition of a single melody of the temporal data set of the data object. The overall melody reflects data and events in a human detectable manner, for example, the melody can reflect stock market trends and signs.

  The equations and formulas presented herein are for illustrative and exemplary purposes only. Various alternative equations and formulas will be apparent to those skilled in the art. Those alternative equations and formulas are within the scope of the disclosed subject matter.

  Embodiments of the system and method of the present invention or portions thereof include computer hardware, firmware and / or storage media (including volatile and non-volatile memory and / or storage elements) each readable by the processor and processor. Can be implemented in a computer program that is executed on a programmable computer or a server. Any computer program can be implemented in a high level procedural or object oriented programming language to communicate with the inside and outside of the computer based system.

  Any computer program may be a storage medium readable by a general purpose or special purpose programmable computer to configure and operate the computer when the storage medium or device is read by the computer to perform the functions of the embodiments (e.g., CD-ROM, hard disk or magnetic diskette) or device (eg, computer peripherals). Embodiments or portions thereof are also implemented as a machine-readable storage medium configured by a computer program that, when executed, causes a machine program to cause a machine to perform the functions of the embodiments described above. Can do.

  The above-described system and method embodiments of the present invention, or portions thereof, can be used in a variety of applications. The embodiments or parts thereof are not limited to the embodiments in this respect, or some of them are microcontrollers, general purpose microprocessors, digital signal processors (DSPs), reductions, among other electronic components. It can be implemented by memory devices in instruction set computing (RISC) and complex instruction set computing (CISC). Also, the above-described embodiments, or portions thereof, may also store main data, cache memory, or other types of data that store electronic instructions that are executed by a microprocessor or that can be used in arithmetic operations. It can be implemented using integrated circuit blocks called memory.

  The description is applicable in any computing or processing environment. Embodiments or portions thereof may be implemented in hardware, software, or a combination of the two. For example, embodiments or portions thereof may be implemented using circuitry such as one or more of programmable logic (eg, ASIC), logic gates, processors, and memory.

  Various modifications to the disclosed embodiments will be apparent to those skilled in the art and the general principles described below may be applied to other embodiments and applications. Therefore, the present invention is not limited to the embodiments shown or described herein.

Claims (62)

For determining the occurrence of the data set for a baseline standard presented as the global sphere that is segmented according to a predetermined rule for separating the data set for placement in a segmented portion of the global sphere In computer-implemented visual analysis methods,
(A) a computer receives input from one or more data feeds to determine a radius of the global sphere that serves as a baseline standard measure by which the occurrence of the data set is measured; Creating a three-dimensional global sphere according to a predetermined rule and displaying it on a display electronically connected to the computer;
(B) the computer segments the global sphere according to a longitude portion to define a first criterion for placing a data point in one of the longitude segments;
(C) the computer segmenting the global sphere according to the latitude segment to define a second criterion for placing the data point in one of the latitude segments;
(D) determining the position of each data point relative to the global sphere according to steps (B) and (C) from information about each data point received by the computer from the one or more data feeds; When,
(E) At time T ₀ , the computer as a three-dimensional sphere having a radius determined by a second predetermined rule for determining the radius for the global sphere and at least by a first performance criterion Representing each data point in a surface position for the global sphere determined according to (D);
(F) the computer tracks the occurrence of each data point from time T ₀ to time T _N according to information about each data point received from the one or more data feeds;
(1) The length of the radius of the three-dimensional sphere associated with the data point according to the second predetermined rule for the computer to determine the radius for the data point in step (E) Changing the first performance criteria in a first predetermined manner when the performance of the data point exceeds the global sphere baseline standard indicator;
(2) The computer determines the radius for the data point in step (E), and the second predetermined method determines if the radius of the data point is less than the global sphere baseline standard index. Changing the length of the radius of the three-dimensional sphere associated with the data point according to the second predetermined rule to change a performance criterion of 1;
(3) The computer determines a radius for the data point in step (E) and maintains the current state of the first performance criterion when the performance of the data point does not change. Displaying on the display one of maintaining a length of the radius of the three-dimensional sphere associated with the data point according to a predetermined rule of two;
(G) selecting a data point according to the occurrence of the data point from T ₀ to T _N.   The method of claim 1, wherein the first criterion is selected from the group comprising color and sound.   The method of claim 2, wherein the first performance criteria in step (E) includes a first color.   The method of claim 2, wherein the first performance criteria in step (F) (1) includes a second color.   The method of claim 2, wherein the first performance criteria in step (F) (2) includes a third color.   The method of claim 1, wherein the global sphere baseline standard indicator comprises a financial indicator.   The method of claim 6, wherein the financial indicator comprises an S & P 500 indicator.   The method of claim 1, wherein each data point comprises representing an individual mutual fund. In a computer-implemented visual analysis method for determining the occurrence of a dataset,
(A) the computer receives input from one or more data feeds and creates a virtual three-dimensional global sphere;
(B) the computer segmenting the virtual three-dimensional global sphere according to a longitude portion to define a first criterion for placing a data point in one of the longitude segments;
(C) the computer segments the virtual three-dimensional global sphere according to the latitude segment to define a second criterion for placing the data point in one of the latitude segments. When,
(D) determining the position of each data point relative to the virtual three-dimensional global sphere according to steps (B) and (C) from information about each data point received by the computer from the one or more data feeds; A step to determine;
(E) for the virtual three-dimensional global sphere determined according to step (D) as a three-dimensional sphere having a radius determined according to predetermined rules and at least according to a first performance criterion at time T ₀ Representing each data point at the surface position of
(F) the computer tracks the occurrence of each data point from time T ₀ to time T _N according to information about each data point received from the one or more data feeds;
(1) The computer determines the radius for the data point in step (E), and the first predetermined method is used when the performance of the data point exceeds the global sphere baseline standard index. Changing the length of the radius of the three-dimensional sphere associated with the data point according to the predetermined rule to change one performance criterion;
(2) The computer determines a radius for the data point in step (E), and the first predetermined method in a second predetermined manner if the performance of the data point is less than the global sphere baseline standard indicator. Changing the length of the radius of the three-dimensional sphere associated with the data point according to the predetermined rule to change the performance criteria of
(3) The computer determines a radius for the data point in step (E) and maintains the current state of the first performance criterion when the performance of the data point does not change Displaying on the display one of maintaining a length of the radius of the three-dimensional sphere associated with the data point according to the rules of:
(G) selecting a data point according to the occurrence of the data point from T ₀ to T _N.   The method of claim 9, wherein the first criterion is selected from the group comprising color and sound.   The method of claim 10, wherein the first performance criteria in step (E) includes a first color.   The method of claim 10, wherein the first performance criteria in step (F) (1) includes a second color.   The method of claim 10, wherein the first performance criteria in step (F) (2) includes a third color.   The method of claim 9, wherein the global sphere baseline standard indicator comprises a financial indicator.   The method of claim 14, wherein the financial indicator comprises an S & P 500 indicator.   The method of claim 9, wherein each data point comprises representing an individual mutual fund. For determining the occurrence of the data set for a baseline standard presented as the global sphere segmented according to a predetermined rule for separating the data set for placement in a segmented portion of the global sphere In computer-implemented visual analysis methods,
(A) The computer receives input from one or more data feeds, thereby determining the radius of the global sphere that serves as a baseline standard indicator from which the occurrence of the data set will be measured. Creating the three-dimensional global sphere having a global sphere center according to a first predetermined rule and displaying it on a display electronically connected to the computer;
(B) the computer segments the global sphere according to a longitude portion to define a first criterion for placing a data point in one of the longitude segments;
(C) the computer segmenting the global sphere according to the latitude segment to define a second criterion for placing the data point in one of the latitude segments;
(D) determining the position of each data point relative to the global sphere according to steps (B) and (C) from information about each data point received by the computer from the one or more data feeds; When,
(E) At time T ₀ , the computer is determined according to step (D) as a three-dimensional sphere having a distance from the global sphere center determined by a second predetermined rule and at least by a first performance criterion. Representing each data point at a surface location for the global sphere;
(F) the computer tracks the occurrence of each data point from T ₀ to T _N according to information about each data point received from the one or more data feeds;
(1) The computer determines the distance from the global sphere center for the data point in step (E), and if the performance of the data point exceeds the global sphere baseline standard index, the first Changing the distance from the global sphere center of the three-dimensional sphere associated with the data point according to the second predetermined rule to change the first performance criterion in a predetermined manner; ,
(2) the computer determines the distance from the global sphere center for the data point in step (E), and if the performance of the data point is less than the global sphere baseline standard index, the second Changing the distance of the three-dimensional sphere associated with the data point from the global sphere center according to the second predetermined rule to change the first performance criterion in a predetermined manner; When,
(3) The computer determines a radius for the data point in step (E) and maintains the current state of the first performance criterion when the performance of the data point is not changed. Displaying on the display one of maintaining the distance of the three-dimensional sphere associated with the data point from the global sphere center according to a predetermined rule of:
(G) selecting a data point according to the occurrence of the data point from T ₀ to T _N.   The method of claim 17, wherein the first criterion is selected from the group comprising color and sound.   The method of claim 18, wherein the first performance criteria in step (E) includes a first color.   The method of claim 18, wherein the first performance criteria in step (F) (1) includes a second color.   The method of claim 18, wherein the first performance criteria in step (F) (2) includes a third color.   The method of claim 17, wherein the global sphere baseline standard indicator comprises a financial indicator.   23. The method of claim 22, wherein the financial indicator comprises an S & P 500 indicator.   The method of claim 17, comprising each data point representing an individual mutual fund. In a computer-based method for interactive analysis of multidimensional data stored in a database,
(A) In a computer-based device, receiving multidimensional data from the database, wherein the multidimensional data is characterized by data values associated with at least a name, a type, and at least a temporal point; Dimensional data has the same name associated with the same entity and the multi-dimensional data has the same type associated with the same category;
(B) providing a user interface for displaying interactive analysis of the multidimensional data on a computer display electronically connected to the computer-based device, the user interface displaying a virtual three-dimensional space; Steps,
(C) on a computer display connected to the computer-based device, displaying a baseline sphere in the virtual three-dimensional space generated by the computer based on the multidimensional data received from the database; The radius of the baseline sphere from the center of the line sphere corresponds to a first index, and the baseline sphere can be segmented by longitude and latitude portions;
(D) In the computer-based device, for each multidimensional data, a corresponding longitude direction and latitude direction coordinate value is calculated, and the longitude direction and latitude direction coordinate values of each multidimensional data are the longitude direction coordinates. Calculated based on the multidimensional data name that defines and the associated multidimensional data type that defines the latitude coordinate;
(E) In the computer display connected to the computer base device, for each multidimensional data, displaying a sphere corresponding to each multidimensional data, and each longitude calculated in step (D) The radius of the sphere arranged in the virtual three-dimensional space according to the direction and latitude coordinate values and associated with each multidimensional data is calculated based on the data value for the multidimensional data, A distance from a center to the center of the baseline sphere is calculated based on data values for the multidimensional data;
(F) selecting multidimensional data by the computer-based device according to its position relative to the baseline sphere.   26. The computer-based method of claim 25, wherein the multi-dimensional data spheres having the same name are displayed in the same longitude portion.   26. The computer-based method of claim 25, comprising displaying the multi-dimensional data spheres having the same type in the same latitude portion.   When the data value of the multidimensional data is larger than the first index, the sphere of the multidimensional data is displayed outside the baseline sphere, and the data value of the multidimensional data is The multi-dimensional data sphere is displayed within the baseline sphere when less than one index, and the multi-dimensional data value is equal to the first index, the multi-dimensional data sphere is displayed inside the baseline sphere. 26. The computer-based method of claim 25, comprising displaying a dimensional data sphere on the baseline sphere.   When the data value of the multidimensional data is greater than the first index, the sphere of the multidimensional data is displayed in a first color, and the data value of the multidimensional data is the first value. The multidimensional data sphere is displayed in a second color when less than the index, and the multidimensional data sphere is equal to the first index when the data value of the multidimensional data is equal to the first index 26. The computer-based method of claim 25, including displaying in a third color.   26. The computer-based method of claim 25, wherein the user interface further includes displaying an interactive timeline corresponding to a time range.   31. The computer of claim 30, comprising displaying a sphere of the multidimensional data in the virtual three-dimensional space when a time value of the multidimensional data corresponds to a selected time value of the timeline. Base method. For determining the occurrence of the data set for a baseline standard presented as the global sphere that is segmented according to a predetermined rule for separating the data set for placement in a segmented portion of the global sphere In the visual analysis system,
Display,
Electronically connected to the display; and
(A) receiving first input from one or more data feeds, thereby determining a radius of the global sphere that serves as a baseline standard measure for which the occurrence of the data set is to be measured; A three-dimensional global sphere is created and displayed on the display according to a predetermined rule.
(B) segmenting the global sphere according to a longitude portion to define a first criterion for placing a data point in one of the longitude segments;
(C) segmenting the global sphere according to the latitude segment to define a second criterion for placing the data point in one of the latitude segments;
(D) determining the position of each data point relative to the global sphere according to steps (B) and (C) from information about each data point received by the computer from the one or more data feeds;
(E) At time T ₀ , in step (D) as a three-dimensional sphere having a radius determined by a second predetermined rule for determining the radius for the global sphere and at least by a first performance criterion Thus representing each data point at the surface location for the determined global sphere,
(F) tracking the occurrence of each data point from time T ₀ to time T _N according to information about each data point received from the one or more data feeds;
(1) changing the length of the radius of the three-dimensional sphere associated with the data point according to the second predetermined rule for determining the radius for the data point in step (E); Changing the first performance criteria in a first predetermined manner when the performance of the data point exceeds the global sphere baseline standard indicator;
(2) determining a radius for the data point in step (E) and determining the first performance criterion in a second predetermined manner if the performance of the data point is less than the global sphere baseline standard index; Changing the length of the radius of the three-dimensional sphere associated with the data point according to the second predetermined rule to change;
(3) determining a radius for the data point in step (E) and maintaining the current state of the first performance criterion when the performance of the data point does not change, One of maintaining the radius length of the three-dimensional sphere associated with the data point according to a rule is displayed on the display;
(G) a system comprising a computer configured to allow a user to select a data point according to the occurrence of the data point from T ₀ to T _N.   35. The system of claim 32, wherein the first criteria is selected from the group comprising color and sound.   34. The system of claim 33, wherein the first performance criteria in step (E) includes a first color.   34. The system of claim 33, wherein the first performance criteria in step (F) (1) includes a second color.   34. The system of claim 33, wherein the first performance criteria in step (F) (2) includes a third color.   35. The system of claim 32, wherein the global sphere baseline standard indicator comprises a financial indicator.   38. The system of claim 37, wherein the financial indicator comprises an S & P 500 indicator.   The system of claim 32, wherein each data point includes representing an individual mutual fund. In a computer-implemented visual analysis system for determining the occurrence of a dataset,
Display,
Electronically connected to the display; and
(A) Receive input from one or more data feeds, create a virtual 3D global sphere,
(B) segmenting the virtual three-dimensional global sphere according to a longitude portion to define a first reference for placing data points in one of the longitude segments;
(C) segmenting the virtual three-dimensional global sphere according to the latitude segment to define a second criterion for placing the data point in one of the latitude segments;
(D) determining the position of each data point relative to the virtual three-dimensional global sphere according to steps (B) and (C) from information about each data point received by the computer from the one or more data feeds;
(E) at a surface position for the virtual three-dimensional global sphere determined according to step (D) as a three-dimensional sphere having a radius determined at a predetermined rule and at least according to a first performance criterion at time T ₀ Represents each data point,
(F) tracking the occurrence of each data point from time T ₀ to time T _N according to information about each data point received from the one or more data feeds;
(1) determining the radius for the data point in step (E), and if the performance of the data point exceeds the global sphere baseline standard index, the first performance criterion in a first predetermined manner; Changing the length of the radius of the three-dimensional sphere associated with the data point according to the predetermined rule;
(2) determining a radius for the data point in step (E) and determining the first performance criterion in a second predetermined manner if the performance of the data point is less than the global sphere baseline standard index; Changing the length of the radius of the three-dimensional sphere associated with the data point according to the predetermined rule to change;
(3) Determine a radius for the data point in step (E) and follow the predetermined rule to maintain the current state of the first performance criterion when the performance of the data point does not change One of maintaining the length of the radius of the three-dimensional sphere associated with the data point is displayed on the display;
(G) a system comprising a computer configured to allow a user to select a data point according to the occurrence of the data point from T ₀ to T _N.   41. The system of claim 40, wherein the first criterion is selected from the group comprising color and sound.   42. The system of claim 41, wherein the first performance criteria in step (E) includes a first color.   42. The system of claim 41, wherein the first performance criteria in step (F) (1) includes a second color.   42. The system of claim 41, wherein the first performance criteria in step (F) (2) includes a third color.   41. The system of claim 40, wherein the global sphere baseline standard indicator comprises a financial indicator.   46. The system of claim 45, wherein the financial indicator comprises an S & P 500 indicator.   46. The system of claim 45, wherein each data point includes representing an individual mutual fund. For determining the occurrence of the data set for a baseline standard presented as the global sphere segmented according to a predetermined rule for separating the data set for placement in a segmented portion of the global sphere In the visual analysis system,
Display,
Electronically connected to the display; and
(A) receiving first input from one or more data feeds, thereby determining a radius of the global sphere that serves as a baseline standard measure for which the occurrence of the data set is to be measured; Creating a three-dimensional global sphere having a global sphere center according to a predetermined rule and displaying it on the display electronically connected to the computer;
(B) segmenting the global sphere according to a longitude portion to define a first criterion for placing a data point in one of the longitude segments;
(C) segmenting the global sphere according to the latitude segment to define a second criterion for placing the data point in one of the latitude segments;
(D) determining the position of each data point relative to the global sphere according to steps (B) and (C) from information about each data point received by the computer from the one or more data feeds;
(E) the global determined according to step (D) at time T ₀ as a three-dimensional sphere having a distance from the global sphere center determined according to a second predetermined rule and at least according to a first performance criterion Represents each data point at the surface location for the sphere,
(F) tracking the occurrence of each data point from T ₀ to T _N according to information about each data point received from the one or more data feeds, and for each data point:
(1) Determine the distance from the global sphere center for the data point in step (E), and if the performance of the data point exceeds the global sphere baseline standard indicator, Changing the distance of the three-dimensional sphere associated with the data point from the global sphere center according to the second predetermined rule to change the first performance criterion;
(2) determining the distance from the global sphere center for the data point in step (E), and if the performance of the data point is less than the global sphere baseline standard indicator, Changing the distance of the three-dimensional sphere associated with the data point from the global sphere center according to the second predetermined rule to change the first performance criterion;
(3) determining a radius for the data point in step (E) and maintaining the current state of the first performance criterion when the performance of the data point is not changed; And maintaining one of the three-dimensional spheres associated with the data point from the global sphere center according to
(G) a system comprising a computer configured to allow a user to select a data point according to the occurrence of the data point from T ₀ to T _N.   49. The system of claim 48, wherein the first criteria is selected from the group comprising color and sound.   50. The system of claim 49, wherein the first performance criteria in step (E) includes a first color.   50. The system of claim 49, wherein the first performance criteria in step (F) (1) includes a second color.   50. The system of claim 49, wherein the first performance criteria in step (F) (2) includes a third color.   49. The system of claim 48, wherein the global sphere baseline standard indicator comprises a financial indicator.   54. The system of claim 53, wherein the financial indicator comprises an S & P 500 indicator.   49. The system of claim 48, wherein each data point includes representing an individual mutual fund. In a system for interactive analysis of multidimensional data stored in a database,
Display,
Electronically connected to the display; and
(A) receiving the multidimensional data from the database, wherein the multidimensional data is characterized by data values associated with at least a name, a type, and at least a temporal point, and the multidimensional data is in the same entity Having the same name associated with the multidimensional data having the same type associated with the same category;
(B) providing a user interface for displaying an interactive analysis of the multidimensional data, the user interface displaying a virtual three-dimensional space;
(C) A baseline sphere is displayed on the display in the virtual three-dimensional space generated by the computer based on the multidimensional data received from the database, and the radius of the baseline sphere from the center of the virtual baseline sphere is Corresponding to a first indicator, the baseline sphere can be segmented by longitude and latitude portions,
(D) For each multidimensional data, calculate corresponding longitude direction and latitude direction coordinate values, and the longitude direction and latitude direction coordinate values of each multidimensional data define the multidimensional data name defining the longitude direction coordinates and Calculated based on the associated multidimensional data type defining the latitude coordinate;
(E) For each multidimensional data, a sphere corresponding to each multidimensional data is displayed on the display, and each sphere is stored in the virtual three-dimensional space according to the longitude direction and latitude direction coordinate values calculated in (D). The radius of the sphere arranged and associated with each multidimensional data is calculated based on the data value for the multidimensional data, and the distance from the center of each multidimensional data sphere to the center of the baseline sphere is the Calculated based on data values for dimensional data,
(F) a system comprising a computer configured to allow a user to select multidimensional data by the computer-based device according to its position relative to the baseline sphere.   57. The system of claim 56, comprising displaying the multi-dimensional data spheres having the same name in the same longitude portion.   57. The system of claim 56, comprising displaying the multi-dimensional data spheres having the same type in the same latitude portion.   When the data value of the multidimensional data is larger than the first index, the sphere of the multidimensional data is displayed outside the baseline sphere, and the data value of the multidimensional data is The multi-dimensional data sphere is displayed within the baseline sphere when less than one index, and the multi-dimensional data value is equal to the first index, the multi-dimensional data sphere is displayed inside the baseline sphere. 57. The system of claim 56, comprising displaying a dimensional data sphere on the baseline sphere.   When the data value of the multidimensional data is greater than the first index, the sphere of the multidimensional data is displayed in a first color, and the data value of the multidimensional data is the first value. The multidimensional data sphere is displayed in a second color when less than the index, and the multidimensional data sphere is equal to the first index when the data value of the multidimensional data is equal to the first index 57. The system of claim 56, comprising: being displayed in a third color.   57. The system of claim 56, wherein the user interface further includes displaying an interactive timeline corresponding to a time range.   62. The system of claim 61, wherein a sphere of the multidimensional data is displayed in the virtual three-dimensional space when a time value of the multidimensional data corresponds to a selected time value of the timeline. .

Images