JP2002539536A - Method and system for accessing clinical information - Google Patents

Abstract

(57) [Summary] A system for promoting a web-based information retrieval and display system is disclosed. Wireless telephones or similar handheld wireless devices with Internet Protocol capabilities are integrated with other peripherals to provide a portable portal to the Internet. The wireless device prompts the user to enter information of interest to the user. With this information, the query is transmitted to the service routine (executed on the web server). This service routine then queries the Web looking for price, delivery, and availability information from various Web suppliers. This information can then be used by nutrition and health monitoring systems to support ongoing exercise programs. The system provides tools for dynamic configuration of mobile communication networks useful in a mobile environment.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD OF THE INVENTION

The present invention relates to an agent-based system, and more particularly, to a mobile computing environment that accesses the Internet to obtain product information required by a user and provides tools for dynamic configuration of a mobile communication network.

[0002]

Problems to be solved by the prior art and the invention

Computer assistance in all settings is becoming essential as computer technology becomes increasingly integrated into society. Mobile computing technology addresses this problem by allowing individuals to access computer-related information at any time and in any environment.

[0003] One of the first major advances in mobile computer technology is the personal digital assistant (PDA). PDAs allow users to access computer related information and are also palm-sized. Using a PDA, a user can organize personal affairs, take notes, calculate equations, and record contact telephone numbers in an address book. In addition, PDAs are typically capable of communicating with desktop computers, which is typically done via a wired connection. With this connection, P
The DA can download information from and upload information to a desktop computer. Subsequent developments have provided wireless capabilities to PDAs. Wireless capabilities have allowed PDAs to interact with other computers that are not physically connected to the PDA.

[0004] Wireless PDAs have become able to communicate with computers connected to the World Wide Web, which has shortly resulted in Web browsable PDAs. One of the first companies to develop PDA's Web browsing capabilities was Intercom. P with Web function by Falcon Mobile Server of Intercom
The DA can be directly connected to the host computer. By simply installing software on the host server, PDA terminals can now access information through the World Wide Web.

[0005] At present, further integration in mobile computing is desired.
Nokia, an Irving, Texas company, partially addressed this integration problem by developing the Nokia 9000 wireless voice telephone. Nokia 90
00 includes a miniature keyboard, a special web browser from Unwired Planet, a microbrowser vendor, and a miniature VGA monitor. Nokia has worked with Ericsson, Motorola and Unwired Planet to establish Wireless Application Protocol (WAP), a standardized browser technology and server format. WAP provides manufacturers with a standardized way to integrate data capabilities into wireless telephones, which allows carriers to increase the degree of control over wireless communication. For example, if a carrier wants to do a field test of a new data service, the carrier performs the service on a server, communicates it to the phone through a microbrowser, and finds that the service is unsatisfactory. If so, you can adjust this service.

FIG. 1A of the prior art is a diagram of a prior art mobile computing solution based on a web portal network. In the prior art, user 1
0 needed to address each participant in the network individually. In a prior art mobile computing solution, a user 10 utilizes an Internet Service Provider (ISP) 12 to create a web portal 14
Get access to The web portal 14 accesses a third party service 16 that provides information directly to the user 10. However, in addition to dealing with the Internet service provider 12, the user 10 needs to purchase a wireless device from a device manufacturer or retailer 18. In most cases, user 1
0 also requires the purchase of a browser from the browser provider 20. Generally, the user will need to pay for wireless communication, which creates a need for the user to transact with the telephone company 22. Finally, the user 10
Needs to do business with the credit card company 24. Clearly, a unified packaged service is the ideal mobile computing solution. Furthermore, a unified package service using agents is highly desirable.

[0007] Agent-based technology has become increasingly important for use in applications designed to interact with a user to perform various computer-based tasks in foreground and background modes. The agent software comprises computer programs set up from the user's perspective to perform routine, redundant, and time-consuming tasks. Agents need to be personalized to suit the goals, habits, and preferences of each individual user, as enabled by the individual user. Therefore, there is a great requirement for an agent in order to efficiently and effectively acquire user-specific knowledge from a user and execute the task on behalf of the user using the knowledge.

[0008] The concept of an agency or a user of an agent is well established. An agent is a person authorized to act on behalf of a party, usually by another party called a party. This allows the party to authorize the agent to perform tasks that the party does not want or cannot perform. For example, an insurance agent can handle insurance requirements for a party, or a talent agent can act on behalf of a performer and arrange a concert schedule.

[0009] With the advent of computers, a new area using agents has emerged.
With significant advances in the field of expert systems, computer programs can act on behalf of a computer user and perform routine, redundant, and other time-consuming tasks. These computer programs are called "software agents".

[0010] Furthermore, recently, computers and communication networks have proliferated. Such networks allow users to access vast amounts of information and services, essentially without any geographic restrictions. Thus, software agents have a rich environment for performing many tasks on behalf of a user. For example, it is now possible for an agent to book an airplane, purchase a ticket, and send the ticket directly to the user.
Similarly, agents can scan the Internet for information from the latest sports information or news to specific graduation theses in applied physics. With current solutions, it is not possible to apply agent technology to obtain information about the upcoming event of the user in a targeted manner.

[0011]

Summary of the Invention

A system for facilitating a web-based information retrieval and display system is disclosed. Wireless telephones or similar handheld wireless devices with Internet Protocol capabilities are integrated with other peripherals to provide a portable portal to the Internet. The wireless device prompts the user to enter information of interest to the user. With this information, a query is transmitted to the service routine (executed on the Web server). This service routine then proceeds to We
b. Query for information on price, delivery, and availability from various web suppliers. This information is formatted and displayed on the screen of the handheld device. The user can then interactively place orders using the handheld device. The system provides tools for dynamic configuration of mobile communication networks useful in a mobile environment.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION

The above and other objects, aspects, and advantages will be better understood from the following detailed description of the preferred embodiments of the present invention with reference to the drawings.

A preferred embodiment of the system according to the invention is preferably implemented in the context of a personal computer, such as an IBM compatible personal computer, an Apple Macintosh computer or a UNIX base workstation. A representative hardware environment is depicted in FIG. 1, where a preferred implementation having a central processing unit 110, such as a microprocessor, and a number of other units interconnected by a system bus 112, is illustrated. 1 illustrates a typical hardware configuration of a workstation according to a form. The workstation shown in FIG. 1 includes a random access memory (RAM) 114, a read-only memory (ROM) 116, an input / output adapter 118 for connecting peripheral devices such as a disk storage unit 120 to the bus 112, A user interface adapter 122 for connecting a keyboard 124, a mouse 126, a speaker 128, a microphone 132, and / or other user interface devices such as a touch screen (no display) to the bus 112, and a workstation through a communication network. (For example, a data processing network) and a display adapter 136 for connecting the bus 112 to the display device 138. The workstation typically resides with an operating system, such as the Microsoft Windows NT or Windows / 95 operating system (OS), the IBM OS / 2 operating system, the MAC OS®, or the UNIX operating system. Those skilled in the art will understand that the present invention can be implemented on such different platforms and operating systems.

[0014] The preferred embodiment uses the Java, C, and C ++ languages and is written using an object-oriented programming methodology. Object-oriented programming (OOP) is increasingly being used to develop complex applications. As OOP becomes mainstream in software design and development, various software solutions require adaptation to take advantage of OOP benefits. There is a need to adapt these OOP principles to the messaging interface of an electronic messaging system so that a set of OOP classes and objects for the messaging interface can be provided.

OOP is a process of developing computer software that uses objects, and includes the steps of analyzing a problem, designing a system, and building a program. An object is a software package that contains both data and a collection of related structures and procedures. Because it includes both data and a collection of related structures and procedures, it can be imagined as a self-sufficient component that does not require other additional structures, procedures, or data to perform a particular task. As such, OOP views a computer program as a nearly self-contained set of components, called objects, each of which is responsible for a particular task. One such data, structure and procedure
The concept of packaging into components or modules is called encapsulation.

In general, an OOP component is a reusable software module that presents an interface consistent with the object model and is accessed at runtime through a component integration architecture. A component integration architecture is a set of architectural mechanisms that allow software modules in different processing spaces to utilize each other's capabilities or functions.
This is commonly done by assuming a common component object model that builds this architecture.

In this regard, it is worth distinguishing between objects and classes of objects. An object is a single instance of a class of an object, often referred to simply as a class. A class of objects can be thought of as a blueprint, from which many objects can be formed.

OOP allows a programmer to create an object that is part of another object. For example, an object representing a piston engine can be said to have a complex relationship with an object representing a piston. In fact, a piston engine comprises a piston, a valve and many other components, and the fact that a piston is an element of a piston engine is logically and semantically represented in OOP by two objects. be able to.

In OOP, an object that “hangs” with another object can also be created. If there are two objects, one representing a piston engine and the other representing a piston engine in which the piston is made of ceramic, the relationship between the two objects is not a complex relationship. Ceramic piston engines do not constitute piston engines. Rather, it is simply a type of piston engine that has one more restriction than a piston engine, namely that the piston is made of ceramic. In this case, the object representing the ceramic piston engine is called a derived object and inherits all aspects of the object representing the piston engine, with additional restrictions or details. An object representing a ceramic piston engine is "hanging" on an object representing a piston engine. The relationship between these objects is called inheritance.

When an object or class representing a ceramic piston engine inherits all aspects of the object representing a piston engine, it inherits the standard piston thermal characteristics defined in the piston engine class. However, ceramic piston engine objects override ceramic-specific thermal properties, which are usually different from those associated with metal pistons. Skip the original and use new functions related to ceramic pistons. Different types of piston engines have different properties, but may have the same basic function associated with a piston engine (e.g., the number of pistons in the engine, ignition sequence,
Lubrication etc.). To access each of these functions in any piston engine object, the programmer calls the same function with the same name, but each type of piston engine implements a different / overridden function implementation behind the same name. May have. Such an ability to hide the implementation of different functions behind the same name is called polymorphism, and this greatly simplifies communication between objects.

With the concepts of compound relationships, encapsulation, inheritance, and polymorphism, objects can represent almost anything in the real world. In fact, the real logical perception we have is the only limitation on determining what types of objects can be objects in object-oriented software. Some typical categories are: ● Objects represent physical objects, such as cars in traffic simulation, electrical components in circuit design programs, countries in economic models, or airplanes in air traffic control programs. Can be. Objects can represent elements of the computer user environment such as windows, menus, or graphic objects. The object can represent a schedule such as a personal file or a table of the latitude and longitude of a city. Objects can represent user-defined data types such as time, angle, and complex numbers, or points in a plane.

OOP allows software developers to design and implement computer programs that are models of some aspects of reality because of the tremendous power of objects that represent almost every logically separable content. And the reality can be a physical entity, process, system, or composite. Because objects can be represented by anything, software developers can create objects that can be used as components in future large software projects.

If 90% of the new software program consists of certified and existing components made from pre-existing reusable objects, only the remaining 10% of the new software project is written from scratch. Need to test. Ninety percent are from extensively tested reusable object evaluations, so the area where errors can occur is 10% of the program. As a result, OOP allows software developers to construct objects from other already constructed objects.

This process is very similar to a complex machine built from assemblies and subassemblies. Therefore, in the OOP technology, a software engineer is close to a hardware engineer in that software is constructed from existing components that can be used as an object by a developer. All of this comes together, leading to improved software quality and increased development speed.

[0025] The programming language is an O.
It has begun to fully support the OP principle. With the advent of the C ++ language, many commercial software developers have adopted OOP. C ++ is an OOP language that provides fast machine executable code. Further, C ++ is suitable for both commercial applications and system programming projects. For now, C ++ is the most popular choice among many OOP programmers, but Smallta
There are many other OOP languages, such as lk, Common Lisp Object System (CLOS), and Eiffel. In addition, the OOP function is
l and more traditional and popular programming languages.

The advantages of the object class can be summarized as follows. Objects and their corresponding classes break a complex programming problem into many small and simple problems. Encapsulation enforces data abstraction through the organization of data into small, independent objects that can communicate with each other. Encapsulation protects the data in an object from unexpected damage, but allows other objects to interact with the data by calling the member functions and structures of the object. • Subclassing and inheritance allows the extension and modification of objects by deriving new kinds of objects from the standard classes available in the system. Thus, a new function is created without having to start from scratch. • Polymorphism and multiple inheritance allow multiple programmers to mix and match properties of many different classes and create special objects that can still cooperate with related objects in a predictable manner. ● Class inheritance and containment inheritance provide a flexible mechanism for modeling real world objects and the relationships between them. ● Reusable class libraries are useful in many situations, but also have some limitations. For example, it is as follows. ● Complexity. In complex systems, class inheritance of related classes can be very confusing with tens or hundreds of classes. ● Control flow. Programs written using the class library are still responsible for the control flow (that is, they need to control the interaction between all objects created from a particular library). The programmer needs to determine which function to call for which kind of object at which timing. ● Duplication of work. Class libraries allow programmers to use and reuse many small pieces of code, but each programmer combines these pieces in different ways. Two programmers can write two programs that do exactly the same thing, using the same set of class libraries, but their internal structure (or design) can be quite different, It depends on many small decisions made by each programmer along the way. Inevitably, similar code snippets will do the same thing in a slightly different way, and will not perform the coordination that should be done.

The class library is very flexible. As programs become more complex, more programmers are forced to reinvent basic solutions to basic problems over and over. An extension of the relatively new class library concept is to have a class library framework. The framework is more complex and consists of a large set of collaborative classes that record small patterns and large mechanisms for implementing common requirements and designs in specific application areas. It was originally developed by free application programmers from the chores involved in displaying menus, windows, dialog boxes, and other standard user interface elements for personal computers.

The framework also represents a change in a programmer's mindset about the interaction of his or her own code with code written by others. Early in procedural programming, programmers called libraries provided by the operating system to perform certain tasks, but basically, the program was executed from start to finish, and only the programmer was allowed to enter the control flow. Responsible. This was suitable for printing paychecks, calculating number tables, or solving other problems with programs that run in only one form.

The development of graphical user interfaces has begun to completely reverse this procedural programming mechanism. Such an interface allows the user to drive the program, rather than the program logic, and determine when certain actions should be performed. Currently, most personal computer software accomplishes this through an event loop that monitors the source of mice, keyboards, and other external events, and calls the appropriate portions of the programmer's code according to the actions performed by the user. Programmers no longer need to determine the order in which events occur. Instead, the program is partitioned into discrete fragments, which are called at unpredictable times and in unpredictable order. By leaving control to the user in this way, the developer creates a program that is easier to use. Nevertheless, individual pieces of a program written by a developer still call libraries provided by the operating system to accomplish a particular task, and the programmer still The control flow within the fragment needs to be determined. The application code is still "at the top" of the system.

Even in the event loop program, the programmer needs to write a large number of codes that need not be individually written for every application. The concept of an application framework goes further than the concept of an event loop. Instead of building the basic menus, windows, and dialog boxes, and then addressing all the nuts and bolts to make them all work together, programmers using the application framework must first create the application code and basic user interface. Place in place. You then build from there by replacing the general capabilities of the framework with the specific capabilities of the intended application.

An application framework reduces the amount of code that a programmer needs to write from scratch. However, because the framework is, in fact, a general application for displaying windows, supporting copy and paste, and so on, programmers can relinquish a greater degree of control than allowed by the event loop program. . The framework code manages almost all event handling and control flow, and the programmer's code is called only when the framework needs it (eg, to create or manipulate its own data structures).

The programmer who writes the framework program not only gives control to the user (the same applies to the event loop program), but also gives the framework a detailed control flow in the program. This approach allows the creation of more complex systems that work together in interesting ways, as opposed to individual programs with custom code created over and over due to similar problems.

Thus, as explained above, a framework is basically a collection of cooperative classes that form a reusable design solution for a given problem area. It typically includes objects that provide default behavior (eg, for menus and windows), and allows the programmer to inherit some of the default behavior and to take other actions such that the framework calls the application code at the appropriate time. Use this by overriding.

There are three main differences between the framework and the class library. ● Behavior versus protocol. A class library is basically a set of actions that can be called when individual actions are required in a program. The framework, on the other hand, provides a protocol or set of rules that governs not only actions, but also the ways in which actions can be integrated, including rules on what the framework provides for what the programmer would provide. . ● Call vs. override. In a class library, a programmer creates an instance of an object by code and calls a member function. Although the framework allows you to create and invoke instances of objects in the same way (that is, treats the framework as a class library), it takes advantage of the reusable design of the framework, and programmers typically override And write the code that is called by the framework. The framework manages the control flow within the object. Writing a program involves separating responsibilities among the various pieces of software invoked by the framework, rather than specifying how the various pieces should operate. ● Implementation versus design. In class libraries, programmers reuse only implementations, while frameworks reuse designs. The framework embodies the way a family of related programs or pieces of software operate. This represents a general design solution that can be adapted to various specific problems in a certain area. For example, a single framework can embody the way a user interface works, but two types of user interfaces created with the same framework can solve completely different interface problems.

Thus, through the development of a framework for solutions to various problems and programming tasks, significant reductions in software design and development work can be achieved. In a preferred embodiment of the present invention, HyperText Markup L
Utilizes anguage (HTML) to implement documents over the Internet, along with a universal secure communication protocol for transporting media between clients and Newco. HTTP or other protocols can easily substitute for HTML without undue experimentation. Information on these products can be found in T. Berners-Lee and D. Connori, "RFC1866: Hypertext Markup Language-2.
0 "(November 1995), and R. Fielding, H. Fristic,
T. Berners-Lee, J. Gettys, and J. C. Mogul's "Hypertext Tran
sfer Protocol-HTTP / 1.1: HTTP Working Group Internet Draft ”(May 1996
On March 2). HTML is a simple data format used to create hypertext documents that can be moved from one format to another. HTML documents are SGML documents with general semantics that are optimal for representing information from a wide range of domains. HTML has been used by the World Wide Web Global Information Business since 1990. HTML is an ISO standard 8879: 1986 information processing text and office system, Standard Generalized Markup Language (S
GML) application.

To date, Web development tools have ranged from client to server,
Limited in the ability to create dynamic Web applications that interoperate with existing computing resources. Until recently, HTML was the dominant technology used in developing web-based solutions. However, HTML has been shown to be unsuitable in the following areas: ● Low performance. ● Limited user interface capabilities. ● Only static Web pages can be created. ● Lack of interoperability with existing applications and data. ● What cannot be standardized.

The Sun Microsystems Java language solves many client-side problems by: ● Improve client-side performance. ● Enable creation of dynamic real-time Web applications. ● Provide the ability to create a wide range of user interface components.

[0038] Java allows developers to provide a robust user interface (UI).
Components can be created. Custom "widgets" (for example,
Real-time stock tickers, animation icons, etc.) can be created, and client-side performance is improved. Unlike HTML, Java supports the concept of client-side validation and offloads appropriate processing to the client for improved performance. Dynamic real-time Web pages can be created. By using the custom UI component, a dynamic Web page can be created.

Sun's Java language has emerged as an industry-recognized language for “programming the Internet”. Sun has described Java (registered trademark) as "simple,
Object-oriented, distributed, interpreted, robust, secure, architecture-neutral, portable, high-performance, multithreaded, dynamic, jargon-compliant, general-purpose programming language " I have. Java supports programming for the Internet in the form of a platform independent Java applet. Java applets are small specialized applications that conform to Sun's Java application programming interface (API), allowing developers to add "interactive content" to web documents (eg, simple animations, pages Decoration, basic game). The applet can copy the code from the server to the client, so that a Java-compatible browser (eg, Netsca
pe Navigator). From a language perspective, the core feature set of Java is based on C ++. Sun's reference to Java (R) states that Java (R) is essentially "C ++ and has an ObjectiveC extension for more dynamic method solutions."

Another technology that provides similar functionality to Java is Microsoft and A
Provided by activeX technology, it gives developers and web designers the tools they need to build dynamic content for the Internet and personal computers. ActiveX includes tools for developing animation, 3D virtual reality, video, and other multimedia content. The tool uses Internet standards, runs on multiple platforms, and is supported by over 100 companies. The building blocks in this group are small, high-speed components called ActiveX controls that allow developers to remove parts of the software from Hypertext Markup Languag.
e (HTML) can be embedded in the page. ActiveX controls work with a variety of programming languages, including Microsoft Visu
al C ++, Borland Delphi, and Microsoft Visual Basic programming system, and in the future, Microsoft's development tool for Java, codename "
Jakarta ”is also included. ActiveX technology is also active
Includes an X server framework that allows developers to create server applications. Those skilled in the art will readily recognize that ActiveX can be substituted for Java® without undue experimentation in order to practice the present invention.

According to a preferred embodiment, as an agent responsible for preparing an individual for an upcoming meeting by helping the individual retrieve relevant information about the meeting from various sources, the background A finder (BF) is implemented. The BF receives the input text in the form of characters indicating the meeting of interest. This input text is generated according to a preferred embodiment by a calendar program containing the time of the meeting. As the meeting time nears, the calendar program is queried to obtain the text of the event of interest, and that information is used as input to the agent. The agent then parses the input meeting text and extracts various components such as title, body, participants, location, time, etc. The system also performs pattern matching to identify particular meeting fields in the meeting text. This information is used to query various sources on the Web to retrieve relevant articles about the current meeting and send them back to the calendaring system. For example, if an individual attends a meeting with Netscape and Microsoft to discuss a dispute between the two companies, the individual will obtain such initial information from the calendaring system. The system then parses the text and recognizes that the meeting company is "Netscape" and "Microsoft" and that the topic is "dispute". The system then queries the Web for relevant information about the topic. Thus, in accordance with the objectives of the present invention, the system updates the calendaring system, ultimately giving the user the best information that can be gathered to prepare the user for the meeting in question. According to a preferred embodiment, this information is stored in a file obtained by selection from a link embedded in the calendar system.

Program Configuration: The computer program according to the preferred embodiment comprises five separate modules:
That is, BF.Main, BF.Parse, Background Finder.Error, BF.PatternMatching,
And BF.Search. In addition, there is a frmMain that provides a user interface that is used only for debugging purposes. Executable programs according to the preferred embodiment should never be executed by this user interface and should only be returned to the calendaring system through Microsoft's Winsock control. The preferred embodiment of the system runs in two modes, which can be specified under the command line sent to the system by the calendaring system. When running in simple mode, the system performs keyword queries and sends them to external search engines. When run in complex mode, the system performs pattern matching before forming a query to send to the search engine.

Data Structures: The system according to the preferred embodiment utilizes three user-defined structures. 1. TMeetingRecord 2. 2. TPatternElement TPatternRecord The user-defined structure TMeetingRecord is used to store any relevant information about a single meeting. This information includes the user ID, the initial description of the meeting, the title of the meeting, and a list of keywords extracted from the body. It is important to note that only one meeting record is created per instance of the system according to the preferred embodiment. This is because each time the system spawns to service an upcoming meeting, it is assigned a task to retrieve information about only one meeting. Therefore, the meeting record created corresponding to the current meeting is examined. ParseMeetingText occupies this meeting record, which is then parsed to provide information about the meeting to other features. GoPa
If tternMatch can bind any value to a particular meeting field, the corresponding entry in the meeting record is also updated. The structure of the tMeetingRecord with the respective fields described in parentheses is presented below according to a preferred embodiment.

A.1.1.1.1.1 Public Type tMeetingRecord sUserID As String (User ID given by Munin) sTitleOrig As String (Original non-stop listed title that needs to be put on the side to send back to Munin) sTitleKW As String ( SBodyKW As String (stoplisted body with only keywords) sCompany () As String (company identified in title or body through pattern matching) sTopic () As String (title or body through pattern matching) SPeople () As String (human specified in title or body through pattern matching) sWhen () As String (time specified in title or body through pattern matching) sWhere () As String ( SLocation As String (location passed by Munin) sTime As String (time passed by Munin) sParticipants () As String (all engaged participants passed by Munin) sMeetingText As String (Original meeting text without user ID) End Type

There are two other structures created to hold each individual pattern used in pattern matching. The record tAPatternRecord is an array containing all the components / elements of the pattern. Type tAPatternEleme
nt is an array of strings representing the elements in the pattern. Since there can be many "alternatives" for each element, an array of strings is needed to keep track of what all the alternatives are. The structures of tAPatternElement and tAPatternRecord are presented below, according to a preferred embodiment.

Public Type tAPatternElement elementArray () As String End Type Public Type tAPatternRecord patternArray () tAPatternElement End Type

Common User-Defined Constants: A number of constants are defined in each declaration section of the program, which may need to be updated periodically as part of the process of maintaining the system according to the preferred embodiment . This constant is accessible to allow for dynamic configuration of the system that occurs as an update to maintain the code.

Included in the table below is a list of constants from each module that I consider most likely to be modified from time to time. However, there are other codes that are not included in the following list but are used in the code. This does not mean that constants not in these lists will never be changed. This means that the frequency of change is small.

Regarding the main module (BF.Main):

[0050]

[Table 1]

Regarding the search module (BF.Search):

[0052]

[Table 2]

About the parse module (BF.Parse):

[0054]

[Table 3]

About the pattern matching module (BFPatternMatch): There are no constants in this module that need to be updated frequently.

General Process Flow: Adjusting the process flows and the functions between them can best be described by the five flowcharts illustrated in FIGS. FIG. 2 illustrates the overall process flow according to the preferred embodiment. Processing begins with a function block 200 at the top of the chart, which is started when the program starts. When the application starts, the command line is parsed, the appropriate meeting text is moved, and the target of the background finding operation according to the preferred embodiment is started, as shown in function block 210. After the targets have been determined, a global stoplist is generated, as shown in function block 220. Thereafter, as shown in function block 230, all patterns utilized in the matching operation are generated. Thereafter, by tracing the chart, function block 200 calls GoBF 240, which is responsible for the logic processing involved in wrapping the correct search query information for the particular target search engine. For example, the function block 240 is a function block 2
Following 50, then call GoPatternMatch as shown in function block 260. To see the GoPatternMatch process flow, swap to the figure titled "Process Flow for BF Pattern Matching Unit".

One important thing to note is that functions represented at the same level of the chart are called sequentially from left to right (top to bottom) by a common parent function. For example, Main 200 calls ProcessCommandLine 210, and then Crea
Call teStopList 220, then call CreatePattern 230, then G
Call oBackgroundFinder240. FIGS. 3 to 6 respectively show the logic of the entire program, the parsing unit, the pattern matching unit, and the search unit in detail. FIG. 6 details the logic that determines the data flow of important information through the BackgroundFinder and shows the functions responsible for creating and processing such information.

Detailed Search Architecture under Simple Query Mode: Search for Alta Vista:

Function Block 270 of FIG. 2 The Alta Vista search engine uses the identification information to return general information about the topic associated with the current meeting, as shown in function block 270 of FIG. The system according to the preferred embodiment retrieves all keywords from the title portion of the original meeting text and uses Alta Vista
Build advanced queries to send to. This keyword is logically integrated in the query. This result is also ranked based on the same set of keywords. One skilled in the art will readily appreciate that data restrictions or publisher criteria can be promoted with respect to the item that one wishes to search for. The set of top ranked articles is returned to the calendaring system according to the preferred embodiment.

NEWS PAGE: (Function Block 275 of FIG. 2) The News Page search system is responsible for providing the latest news topics regarding the target meeting. This system extracts all keywords from the title part of the original meeting text and
s Construct a query to send to the Page search engine. This keyword is logically integrated in the query. Only recently published items are searched.
The News Page search system provides data restriction criteria that can be set by the user according to the user's preference. Top ranked articles are returned to the calendaring system.

FIG. 3 is a user profile data model of the preferred embodiment. Processing begins at function block 300, which is responsible for calling a program from a main module. Thereafter, in a function block 310, a wrapper function is called, and a keyword extraction process in a function block 320 is prepared. After the keywords have been extracted, processing flow continues to function block 330, where it is determined whether the delimiters are correctly placed. Thereafter, at function block 340, the number of words in the particular string is calculated, and at function block 350, the particular field delimiter and the particular field from the meeting text are retrieved. Thereafter, in function block 380, the string delimiters are checked again to ensure proper placement. Finally, in function block 360, the extraction of each word from the title and body of the message is performed one word at a time, and a function block 362 to find the next closest word delimiter in the input phrase, A function block 364 to strip data and determine if the word is in the stoplist;
If the word is on the stoplist, logic is used with a function block 366 that returns an error.

Pattern Matching According to Preferred Embodiment: Restrictions associated with the simple search method include:

1. Relying on stoplists of unwanted words to extract a set of keywords from the meeting text is limited by how comprehensive the stoplist is. Rather than thinking about which parts of the meeting text should be discarded, focus on what parts of the meeting text are needed.

2. The simple search method according to the preferred embodiment uses only the keywords from the meeting title and uses Alta Vista and News Page.
Form a query to send to. It ignores the alternate source of the query, the body of the meeting notification. It is not possible to include keywords from the meeting body in the formation of the query, since often long and complex queries result in meaningless results.

[0065] 3. There is no way to tell what each keyword means. For example, “Andy” and “globe” may be extracted as two keywords. However, with Simple Search, there is no way to know that "Andy Grove" is actually a person's name. Imagine the possibility of somehow intelligently inferring that "Andy Grove" is a personal name. Information about where he is employed and where he currently lives.

[0066] 4. In summary, they rely on stoplists alone to parse and filter out unnecessary words, which plagues them with "too much information."

According to a preferred embodiment, pattern matching overcomes these limitations: Here we show how pattern matching can address each of the corresponding issues above, according to a preferred embodiment.

1. By performing pattern matching, only a necessary part of the meeting text is matched, and this part is extracted.

[0069] 2. Perform pattern matching on the meeting body and extract only the necessary parts from the meeting body. Thus, the meeting body is not completely wasted.

[0070] 3. Pattern matching is based on a set of specified templates,
Allows company names and other items to be identified from the meeting text.

[0071] 4. In summary, pattern matching will not bother you with information overload. The big question, of course, is how well pattern matching works. Relying on artificial intelligence processing alone does not result in a 100% hit rate. It is possible to specify about 20% of the provided company name.

Pattern: A pattern in the context of the preferred embodiment is a template that specifies the structure of the phrase you are looking for in the meeting text. Since a pattern is a template for words that have a high probability of appearing in someone's meeting text,
The pattern supported by the preferred embodiment is selected. For example, when entering a meeting in a calendar, many people say "Meet with Bob Dutton from Stan
ford University next Tuesday (to meet Bob Dutton of Stanford University next Tuesday). " In this case, the common pattern would be the word "with" followed by a person's name (Bob Dutton in this example), followed by the word "from", and ending with an organization name (Stanford University in this example) .

Terms for Pattern Matching: Common terms related to pattern matching are presented below. ◆ Pattern: A pattern is a template that specifies the structure of a phrase that you want to associate with the meeting text. It contains subunits.

Element: A pattern can include many subunits. These subunits are called elements. For example, in the pattern "with $ PEOPLE $ from $ COMPANY $", "with", "$ PEOPLE $", "from", and "$ COMPANY $" are all elements.

◆ Placeholder: A placeholder is a special kind of element that needs to be linked to a value. Using the example above, "$ PEOPLE $" is a placeholder.

Indicators: Indicators are another type of element that need to be found in the meeting text but do not need to be tied to a value. Often a particular pattern looks for one or more indicators. So the indicator is not of the "atomic" type.

◆ Replacement characters: A replacement character is a set of indicators that are all synonymous with each other. It is good to find any of these in the input.

There are five fields specified for each meeting. ◆ Company ($ COMPANY $) ◆ People ($ PEOPLE $) ◆ Location ($ LOCATION $) ◆ Time ($ TIME $) ◆ Topic ($ TOPIC_UPPER $) or ($ TOPIC_All $) The placeholder I used to represent the meeting field.

Each placeholder has the following meaning. ◆ $ COMPANY $: Connects a string of words starting with an uppercase letter (eg, Mee
t with Joe Carter of <Andersen Consulting>).

$ PEOPLE $: Joins a sequence of two words starting with a capital letter that may be joined by ",", "and", or "&" (eg, Meet
with <Joe Carter> of Andersen Consulting, Meet with <Joe Carter and Luk
e Hughes> of Andersen Consulting).

$ LOCATION $: Connects a string of words starting with a capital letter (eg, Me
et Susan at <Palo Alto Square>).

$ TIME $: Format #: Connect a string containing ## (eg, Di
nner at <6:30 pm>).

◆ $ TOPIC_UPPER $: Connects a string of words starting with uppercase letters about a topic (eg, <Stanford Engineering Recruting> Meeting to talk about n
ew hires).

$ TOPIC_All $: Connects a string of words, regardless of whether they start with a capital letter (eg, Meet to talk about <ubiquitous computing>)
.

The following is a table showing all the patterns supported by BF. Each pattern belongs to a pattern group. All patterns in a pattern group share a similar format and differ only in which indicators are used as replacement characters. Note that the gray pattern is also commented out in the code. Although BF has the ability to support these patterns, we have determined that matching these patterns is not essential at this time.

[0086]

[Table 4]

FIG. 4 is a detailed flowchart of pattern matching according to the preferred embodiment. Processing begins at function block 400, where the main program calls the pattern matching application and passes control to a function block 410,
Start the pattern matching process. Next, in function block 420, the wrapper function continues to loop and process each pattern, including determining whether a portion of the text string can be tied to the pattern, as shown in function block 430. . Then, in function block 440, various placeholders, if any, are associated with the values, and in function block 441, a list of names separated by punctuation is associated, and function block 442 is provided.
, The full name is processed by finding the two words starting with an uppercase letter as the full name and capturing the next letter after the space after the word to determine if it starts with an uppercase letter. Next, at function block 443, the time is parsed from the string in an appropriate manner, and at function block 444 the next word after the blank space is parsed. Next, in function block 445, successive words starting with a capital letter, such as company, topic, or location, are combined;
At function block 446, the next word after the blank is obtained for further processing according to the preferred embodiment. Following the match meeting field processing, function block 45 locates the indicator at the beginning of the pattern.
A zero is utilized and the next word after the blank is obtained, as shown in function block 452, and this word is checked, as shown in function block 454, to determine if it is an indicator. Next, in function block 460, the string is parsed to locate the indicator that is not at the end of the pattern, and the next word after unnecessary white space, such as that following a line feed or carriage return, is shown in function block 462. And the word is analyzed to determine if it is an indicator, as shown in function block 464. Next, function block 4
At 70, the temporary record is reset to a null set to prepare for processing of the next string, the meeting record is updated at function block 480, and a check is performed at function block 482 before parsing the meeting record. It is determined whether an entry has already been made in the meeting record.

Using Identified Meeting Fields: Having identified fields in meeting text that we consider important, there are many things we can do with them. One of the most important pattern matching applications, of course, is to improve the queries we have built that ultimately send to Alta Vista and News Page. There are many other options and enhancements that utilize the results of pattern matching that can be added to the BF. These other options are described in the next section. The purpose of this section is to give readers a good understanding of how pattern matching results can be used to assist in obtaining good search results.

FIG. 5 is a flowchart of a detailed process according to a preferred embodiment for preparing a query and obtaining information from the Internet. Processing proceeds to function block 500
And immediately continues to a function block 510 to process a wrapper function for Internet search using a Web search engine. If this search utilizes the Alta Vista search engine, function block 530
At, the system retrieves information from the meeting record and forms a query to send to this search engine in function blocks 540-560.
If the search utilizes the NewsPage search engine, at function block 520, the system retrieves information from the meeting record and forms the query at function blocks 521-528.

Alta Vista Search Engine: An advantage of the Alta Vista search engine is that it offers enhanced flexibility. Using advanced query methods, you can build any kind of Boolean query and rank your search as you wish. However, Alt
One of the biggest drawbacks of a Vista is that it is not good at handling large queries and is more likely to return irrelevant results. If the topic and company can be identified in the meeting text, very short but comprehensive queries can be formed where good results are expected. In addition, we need to pay attention to the topics we have discovered. Finding information about a company may not be a significant benefit to the user, especially if the user is already familiar with the company and has numerous meetings with the company. These users want to research on topics.

News Page Search Engine: The advantage of the News Page search engine is that it does a good job in searching for the latest news, if it is possible to give a valid company name. Therefore, when sending a query to the news page Web site, any identifiable company name is sent, and only when the company name cannot be found, the found topic is used for forming a query. If neither is found, no search is performed. Alta
The algorithm used to form the query to send to Vista is illustrated in FIG. The algorithm used to formulate a query to NewsPage is illustrated in FIG.

The following table details each function according to the preferred embodiment.
The order in which the functions appear is as similar as possible to the process flow. When there is a situation where a function is called several times, this function is described after the first function that calls this function, and the description is not repeated after the subsequent function that calls this function.

[0093]

[Table 5]

FIG. 6 illustrates Alta Vista and News Pa according to a preferred embodiment.
5 is a flowchart of actual codes used for preparing and sending a search for a ge search engine. Processing begins at function block 610, where a command line is utilized to update a calendar entry with specific calendar information.
This message is then posted according to function block 620 and function block 63
A meeting record is created to store the current meeting information according to 0. Next, in function block 640, the query is Alta Vista
Submitted to the search engine, and at function block 650, the query is News
Sent to Page search engine. When the message is returned from the search engine, it is stored in a result data structure as shown in function block 660, and this information is processed as described in function block 670 and summarized in a file used in preparing for the meeting. Stored in format.

FIG. 7 presents details regarding the creation of a query according to a preferred embodiment. Processing commences at function block 710, where the meeting record is parsed to obtain possible company, person, topic, location, and time. Thereafter, at least one topic is identified at function block 720, at least one company name is identified at function block 730, and finally at function block 740, which data is stored in the file for final consumption by the user. A determination is made as to whether to transmit.

FIG. 8 is a variation of the theme of the query presented in FIG. The meeting record is parsed in function block 800 and the company
At 20, the topic is identified at function block 830, and finally, at function block 840, the topic and / or company name is used to formulate the query.

[0097] Alternative embodiments that add various specific features for specific user requirements are described below.

Reinforcement of Target Ratio of Pattern Matching: In order to improve the performance of BF, a pattern / pattern group is added to the procedure “CreatePatterns”. Existing code that declares a pattern can be used as a template for future patterns. Because everything is stored as a dynamic array, it is convenient to reuse code by cutting and pasting. Functions BindName, BindTime, Bind responsible for associating values with placeholders
CompanyLocTopic can be enhanced. This enhancement is achieved by increasing the set of criteria that link a particular meeting field to increase the number of values to link. For example, BindTime currently accepts and binds all values in the form ##: ## or ##: ##. To increase the time that can be combined, the more aesthetic time term "o'clock" can be followed by the digits 1 through 12. By assigning an accuracy rate to each guess made by the vocabulary-based recognition algorithm and the BF, the only guess that satisfies a certain threshold to be valid is possible.

Depending on which location the system has identified through pattern matching, or which location the user has indicated as a meeting location, the system according to the preferred embodiment detects the word lunch / dinner / breakfast. Whenever you do, we suggest several excellent restaurants. Furthermore, using a site such as Company Finder, it is possible to confirm that what is at hand is really the company name, or if the company name cannot be specified by pattern matching, use the company finder Web site as a "dictionary". Can be used to determine whether a particular word starting with a capital letter represents a company name. It is also possible to display the stock price and the emergency report of the specified company.

Wireless Bargain Identification According to a Preferred Embodiment: FIG. 9 is a hardware and logical control flow of equipment and software designed to enable Web-based comparative shopping in a traditional physical non-Web retail environment. FIG. A wireless telephone or similar handheld wireless device 920 with Internet Protocol capability is integrated with a small barcode reader 910 (located inside the telephone or with a short cable) and a Unified Product Code (UPC) for books and other products 900 Used to scan barcodes. The wireless device 920 transmits the barcode to the pocket bargain finder service module (executed by the Web server) 940 via the antenna 930,
This translates the barcode into an international standard book number (for books) or an appropriate identifier (for other products). This service module will then contact the appropriate third party web site (s) and contact the various web suppliers 9
Find price, shipping, and availability information for products from 50. This information is formatted and displayed on the screen of the handheld device. IP wireless telephones and other handheld devices 920 are available from Metricom's Ricochet SE Wirel
Use a wireless modem such as ess Modem. By using this device, users can
While relaxing in a coffee shop, the shaking latte can be dangerously close to the keyboard, and a portable computer on an old small table can access the Internet as fast as a direct telephone line connection.

The 8 oz (227 g) Ricochet SE Wireless Modem is about the size of a pack of cigarettes, is extremely simple to set up, attaches the modem to the back of the portable device with the included Velcro components, and connects the cable to the serial device. Just plug it into a port, wake up a short, thick antenna, and transmit. Software setup is equally easy, with a simple installer adding a Ricochet modem driver and placing a connection icon on the desktop. The functional aspects of the modem are the same as a conventional telephone modem.

Of course, wireless performance is not as reliable as a conventional dial-up connection. In some places in San Francisco, you could get a strong connection as long as you were near the window. However, Ricochet was not able to connect at all in CNET's brick office. When connected online, performance up to 28.8 kbps is available with a slow drop to slower speeds. However, there was no disappointment at slow speeds. Compared with the alternative method-connection by cellular modem-
Ricochet is much faster, more reliable, and less costly to use. Of course, this
SE Wireless is battery powered. The continuous battery life of this modem is up to 12 hours. According to a preferred embodiment, the portable computer's dual batteries run out before the output of the Ricochet diminishes.

Thus, utilizing this wireless modem, the user utilizes the web server software 940 to identify the appropriate product 950 and then the appropriate device (s), according to a preferred embodiment. You can select a supplier using the key and place an order. The bargain finder service module then completes the order with the appropriate third party web supplier 960.

[0104] mySite! Prototype of personal website and intention value network: mySite! Is a high-impact Internet-based application according to the preferred embodiment, focusing on the theme of providing each customer with a personalized experience through a personal website in a buyer-centric world. is there. This service is intended for customers-judgments that require basic living needs or extensive planning and coordination across several dimensions, such as financial planning, health care, personal and professional development, family life, and other issues. It is organized intuitively with a focus on satisfying. Each member has and maintains their own profile, which allows them to create and browse content in systems that target themselves in a limited manner. From the point of entry of a product or service request to the completion of payment, intelligent agents are used to conduct research, execute transactions, and provide advice. By using advanced profiling and filtering, intelligent agents learn about users and improve the services they provide. Customer intent includes daily logistics management (eg, email, calendar, contacts, agendas, bill payments, shopping, and travel planning) and moving to new communities (eg,
Finding new places to live, moving family property, going on trips and sending insurance, notifying work and personal contacts, learning about new communities). From a consumer perspective, mySit
e! Provides a central place where users can access related products and services and accomplish their daily tasks very easily and conveniently.

From a business point of view, mySite! Represents a value-added and innovative way to effectively attract, serve and retain customers. The intention value network allows users to enter personalized sites and interact seamlessly with network participants with the help of intelligent agents learning. The intention value network according to the preferred embodiment provides excellent value. 24 per day for personalized information, advice and products
Time, seven days a week access is provided. Because this information is personalized, each user will browse highly customized content to ensure the relevance to the required target user.

Egocentric Interface: An egocentric interface is a user interface created to satisfy the needs, preferences, and current context of a particular user. It utilizes the user's personal information stored in the central profile database to customize the interface. The user can turn on security permissions and set preferences for interface elements and content. The content integrated into the egocentric interface is customized with relevant information about the user. When displaying content, the egocentric interface includes a relationship between the content and the user in a manner that demonstrates how the content is relevant to the user. For example, when a user displays information about a contracted upcoming ski trip, the interface includes information about events from the user's personal calendar and contact list, such as other people in the same area during the ski trip. included. This serves to put new pieces of information into contexts that the individual user is familiar with.

FIG. 10A illustrates an implementation of an intent value network architecture on the World Wide Web. For the sake of simplicity, this diagram ignores the complexity associated with security, scalability, and privacy. Customer, Netscape N
Any internet Web browser 1010 running on a personal computer connected to the Internet, such as an avigator or Microsoft Internet Explorer, or a portable information terminal having wireless capability can access the intention value network. See FIG. 17 for a detailed description of the multiple ways to access the intention value network. Customer must have a unique name or Integrator We
b. Access the intention value network by the IP address associated with the server 1020. The integrator creates a network of intention values using a combination of resources such as an intention database 1030, a content database 1040, a supplier profile database 1050, and a customer profile database 1060.

The intention database 1030 stores all information regarding the structure of the intention and the types of products and services required to satisfy the intention. The information in this database includes the step of intention, the area of interest, the layout template, and the personalization template. Content database 1
Reference numeral 040 stores all information related to intention, such as advice, introduction information, personalized content, satisfaction evaluation, product evaluation, and progress report.

The supplier profile database 1050 contains information about product and service providers integrated into the intent. The information contained in this database provides a link between the intent framework and the supplier. This includes a product list, features and descriptions, and the address of the supplier's product website. The customer profile database 1060 includes names, addresses, social security numbers and credit card information, personal preferences, behavior information, history,
It includes personal information about the customer, such as Web site layout preferences. Supplier's web server 1070 provides access to any supplier's databases needed to provide information and transactional support to customers.

[0110] The product information database 1080 stores all product-related information such as features, availability, and prices. Product order database 1090 stores orders for any customer. Interfaces to this database are SAP, Baan, O
This can be done through an integrated line-of-business application provided by racle et al., or it can be accessed directly through the supplier's web server or application server. Customer information database 1091 stores any customer information needed by the supplier to complete a transaction or maintain a customer record.

FIG. 10B is a flowchart presenting the logic used to create a Web page within the egocentric interface. In this environment, it is assumed that a Web server and a Web browser are connected by a TCP / IP network such as a public Internet or a private intranet. Possible Web servers include Microsoft Internet Information Server, N
Includes etscape Enterprise Server or Apache. Possible web browsers include Microsoft Internet Explorer or Netscape Navigator. The client (that is, Web browser) sends a message to the server (that is, Web server)
1001 for requesting a specific Web page. This is typically accomplished by a user clicking a button or link in a web page. The Web server 1002 obtains the layout and content preferences for that particular user in response to a request to a user profile database 1003 and a database that manages user IDs stored on the client (ie, Web browser). Next, W
The eb server searches the content database 1005 for the content of the requested page 1004. Next, relevant user-centric content such as calendar, email, contact list, and task list items is searched 1006 (
See FIG. 11 for a detailed description of this process). Queries to the database use the user's content preferences stored as part of the user profile in the user profile database 1003 to filter the returned content. The returned content is then formatted 1007 into Web pages according to the layout preferences defined in the user profile. The web page is then returned to the client and displayed 1008 to the user.

FIG. 11 shows a process of searching for user-centered content to be added to a Web page. This process describes 1006 in FIG. 10B in more detail. It is assumed here that the server has already obtained the user profile and the existing content to be integrated into this page. The server parses the filtered content 1110 for instances of events, contacts, names, and email addresses. If any of these are found, they are tagged and stored in the temporary holding space. Next, the server attempts 1120 to find any user-centric content stored in various databases. To do this, the tagged items in the temporary storage space are stored in the calendar item 1130 of the calendar database 1140, the e-mail item 1115 of the e-mail database 1114, the contact item 1117 of the contact database 1168, and the task list database 1118. This involves matching the task list item 1119 with the news item 1121 of the news database 1120. After searching for relevant user-centric content, this is edited back together 112
22.

User Personas: The system allows a user to create a number of different personas that aggregate profile information into a valid set in different contexts. The user can create one persona when making a purchase for home. The persona may include a home address and may indicate that the user is looking for a good bargain when shopping. The same user can create a second persona that can be used when in the context of work.
This persona may store the location of the user's workplace and may indicate that the user likes a particular vendor or works for a particular company that has set up a discount program. is there. The user can use this persona when shopping for work related items. Personas can also include rules and restrictions. For example, a work persona may restrict a user to book an airplane at only one travel agency and to use the arrangement rules set by the employer.

FIG. 12 shows a relationship between a user, a plurality of personas, and a plurality of profiles. At the user level is a user profile 1200. This profile represents the user and his account information. There is one unique record in the database for each user who has an account. Each user is associated with a plurality of personas 1220, 1230, and 1240. These personas are used to group profiles into valid contexts. For example, suppose a user lives in San Francisco and works in Palo Alto, but owns a mountain cottage on Lake Tahoe. He has three types of contexts that may access his site. One context is work-related. The other two are related to family life, but in different places. The user can create a work persona 1220, a home persona 1230, and a mountain villa persona 1240. Each persona has a different general profile 12
50, 1260, and 1270, including the address of the location.
Therefore, there are three general profiles. Each persona also
Reference one of the two travel profiles. This user maintains a work travel profile 1280, which includes all company rules for arranging tickets and performing bookings. The profile may specify, for example, that this person travels only in business class or first class, and that his favorite airline is United Airlines. The work persona references this work travel profile. The user can further maintain a home travel profile 1290 that specifies that he prefers to travel in economy class and is seeking a non-refundable sale because it is generally cheap. Both the home persona and the villa persona point to the home travel profile.

FIG. 13 shows a data model that supports the concept of persona. User table 1310 includes a record of each user who has an account in the system. This table includes a user name and password 1320 and a unique identifier. Each user can have multiple personas 1330, which serve as containers in a more specialized structure called a profile 1340. The profile contains the detailed personal information in the profile field 1350 record. Each profile comes with a set of profile restrictions 1360 records. Each of these has a name 1370 and a rule 1380
Which defines the limit. The rules are in a pattern-like form (y if x) so that the rules can be restricted to specific uses. One example of a profile restriction is a rule that stipulates that a user cannot book an airplane with a particular airline listed. This profile restriction can be included, for example, in the "travel" profile of the "work" persona set by the user's employer. Each profile field further includes a set of permissions 1390, which are included in the record. These permissions define who has what access to the information in a particular profile field.

Intent-centric interface: To satisfy customer intent, such as retirement or relocation planning, a specialized interface is needed. Customer intent requires extensive planning and coordination across a number of areas, from, for example, financial security, housing, and transportation to healthcare, personal and professional growth, and entertainment. Satisfaction of intent requires a complementary business network that works across industries to help meet consumer needs.

An intention-centric interface is a user interface designed to help a user manage personal intentions. At any given point in time, the content of the interface is customized and only content relevant to these particular intents is shown. The intention-centric interface allows the user to manage the process of satisfying these particular intentions. This involves a series of separate steps and a set of content areas accessible to the user. At any time, the user can switch to an interface that manages different intents, which changes the content of the interface to include only content relevant to satisfying the newly selected intent. Become like

FIG. 14 provides a detailed description of the data model needed to support the intention-centric interface. Each user persona 1410 (see FIG. 13 for a detailed description of the persona data model) has any number of active user intents 1420. Each active user intent is given a nickname 1430, which is the display name that the user sees on the screen. Each active user intent further includes a number of data fields 1440, including any user data collected throughout the user interaction.
For example, if the user fills a form on the screen and one of its fields is a social security number, the corresponding data field is name = "SSN" 1460
, Value = “999−99−9999” 1460. Each user intent further tracks the intent step 1470 completion status. Complete 1480
The field indicates whether the user has completed the step. Every user intent is a user specific version of the general intent 1490, and the general intent is a default model of the intent for every user. General intent is customized through custom rules 1411 and 1412 that accompany the intent sub-steps. These custom rules are patterns that describe how the system uses the profile information of individual users to customize each user's intent.

Statistics Agents: Agents keep track of key statistics for each user. These statistics are used in a manner similar to the virtual reality pet Tamagotchi to prompt the user for specific actions. These recorded statistics are the frequency of logins, the frequency of evaluating content such as news articles, and the activity of the agent as measured by the number of tasks performed during a particular period. This information is used by the system to appeal emotionally to the user for specific actions.

FIG. 15 illustrates the process of generating a page that displays the current statistics of the agent. When the user requests the agent statistics page via the client browser 1510, the server retrieves 1520 the user's statistics from the user's profile database 1530. The server then performs 1540 the mathematical calculations necessary to create a standardized set of statistics. Next, the server searches the content database 1560 for an expression to use to calculate the user-centered statistics 1550. Thereafter, a graph is generated 1570 using the general formula and the statistics of the user. This graph is inserted 1580 into a template for creating a statistics page. The page is then returned 1590 to the user.

Personalized Product Report Service: The system provides services such as consumer reports customized for each user based on the user profile. The system records and provides user ratings of product quality and desirability in many dimensions. The difference between this system and the conventional product quality measurement service is that the evaluation returned to the user is personalized. The service works by finding the person who has the closest match to the user's profile and has previously rated the product being sought. Using this algorithm is
It helps to ensure that a product report is sent back to the user containing only statistics from people similar to the user.

FIG. 16 shows an algorithm for determining a personalized product evaluation for a user. When a user requests a product report for product X 1610, the algorithm retrieves 1620 a profile from a profile database 1630 (including product ratings) of users who have previously rated the product. Next, the system searches the content database 1650 for a default threshold for the profile matching algorithm 1640. Next, map all of the user's candidate lists to several dimensions specified in the profile matching algorithm 1660. The top n closest neighbors are then determined (pre-specified as a threshold variable), and using the results of the profile matching algorithm, they are separated from the profiles of the users in the set by y (also as a threshold variable). Execute a test to determine if it is in (specified in advance) 1
670. If not, the threshold variable is relaxed 1680 and the test is run again. This process is repeated until the test returns true. Thereafter, a number of product statistics along several dimensions are determined 1690 using product ratings from a small set of n nearest neighbors. These statistics are inserted 1695 into the product report template and returned to the user as a product report 1697.

[0122] Uneven distribution of personal profiles and services: This system provides a centralized location for storing personal profiles. This storage location is a server available through the public Internet and is accessible by any device with appropriate access capabilities connected to the Internet. With access to the profile everywhere, a number of access devices can be used to customize services for the user based on the user's profile. For example, a merchant website can use this profile to provide personalized content to users. A mobile information terminal (P
DA) allows personal calendars, emails, contact lists, task lists, and notes on a PDA to be synchronized with versions stored on an Internet site. This allows an individual to maintain only one version of this data in order to keep the data available regardless of the required time and format.

FIG. 17 presents the detailed logic associated with a number of different ways to access such a centrally stored profile. The profile database 1710 is a central storage location for user profile information. The profile gateway server 1720 receives any requests for profile information, whether from the user himself or from a vendor seeking to serve the user. This profile gateway server is responsible for ensuring that information is released only when the profile owner explicitly gives permission. Any device that can access the public Internet 1730 via TCP / IP (Standard Network Communication Protocol) is an intelligent HTT
With the P request, information from the profile database can be requested. Consumers include, for example, televisions 1740, mobile phones, smart cards, gas meters, water meters, kitchen appliances, security systems, desktop computers,
Access to the service can be obtained from devices such as laptops, pocket organizers, PDAs, and automobiles. Similarly, merchants 1750 can also access such profiles (with permission from the consumer who owns each profile), thereby providing consumers with customized and personalized services.

One potential use of this ubiquitous profile relates to hotel chains. Consumers can carry a smart card with a digital certificate that uniquely identifies them. The digital certificate of the smart card was issued by the system, and the system records consumer profile information in a profile database. The consumer takes the card to the hotel chain and checks in. The hotel employer swipes the card, the consumer enters a PIN and unlocks the digital certificate. The certificate is sent to the profile gateway server (using a secure transmission protocol) and is authenticated. This allows the hotel to allow the consumer to access certain portions of the previously specified consumer profile. Hotels can search for all billing information and hotel room preferences of consumers. In addition, hotels can also access consumer movie and dining preferences, and provide customized menus for both. The hotel can provide a service that sends an email to the consumer's spouse to let them know that they have checked in and are safe. Any transaction information can be uploaded to the consumer profile after checking into the hotel. This allows the hotel partners to utilize the information about the consumer collected by the hotel (also with permission from the consumer).

Intention Value Network: In the Intention Value Network, the overall integrator system coordinates the delivery of products and services to users. This integrator manages a network of approved suppliers that physically and virtually provide users with products and services based on user preferences reflected in the user's profile. This integrator manages the relationship between the supplier and the consumer, and coordinates the fulfillment of the consumer's intent by the supplier. Integrators do this by, for example, providing consumers with information about products and suppliers and providing objective advice.

FIG. 18 discloses a detailed interaction between a consumer and an integrator involving one supplier. This user accesses the web browser 1810,
Request product and price information from the integrator. This request is sent from the user's web browser to the integrator's web / application server 1820. User preferences and personal information are obtained from the integrator's customer profile database 1830 and returned to the web / application server. The requested product information is extracted from the supplier's product database 1840 and customized for a particular customer. The Web / application server updates the supplier's customer information database 1850 with the question information about the customer. The product and price information is then formatted on a web page 1860 and the customer's W
It is returned to the web browser.

Summary Agents: A suite of software agents running on applications and web servers is programmed to manage repetitive or routine tasks for users. This agent operates according to the rules set by the user and can only perform tasks explicitly defined by the user. The agent can manage paying users, filtering content and emails, and providing a summary view of tasks and agent activities. The user interface for the agent can be modified for a particular user.

FIG. 19 discloses logic in accordance with a preferred embodiment of processing by an agent that generates an oral summary for a user. When the user requests a summary page 1900, the server obtains 1920 the user's agent preferences, such as agent type, rules, and summary level, from the user profile database 1930. The server obtains 1940 content from the content database 1950, such as email, schedule list items, news, and billing. The agent uses rules stored in the profile database to parse 1960 all such content and create a summary of the content. The content is formatted 1970 into a web page according to the template. An agent speech text is generated 1980 using the content of the content database 1990 and a speech template stored in the database. This speech text is inserted into the web page and
95, the page is returned 1997 to the user.

Trusted Third Party: In the above scenario, the website needs to maintain a guarantee of information privacy according to the published policy. The system is a trusted third party of the consumer, acting on behalf of the consumer in all cases, and is too biased towards information privacy rather than simulation of commercial opportunities. This trusted third party has the appropriate set of processes to guarantee a particular concatenation with the stated policy.

“Me-commerce”: This word is an extension of the word “e-commerce” and means “personalized electronic commerce”. FIG. 20 illustrates a display login according to a preferred embodiment. This display is available for Microsoft with Agent 2000
Implemented as an Internet Explorer application, the agent guides the user through the process of interacting with the system to customize and personalize various system components, gather information, and interact with the user's personal requirements. The user inputs the user name into 2010 and the password into 2020, selects the button 2040, and starts the login procedure. As indicated by the logo 2030, the system converts e-commerce into personalized, so-called "
me ”commerce.

FIG. 21 illustrates a daily logistics management display according to a preferred embodiment. The user is greeted by the animation agent 2100 with a personalized message 2190. The user can travel 2110, household chores 21
Various activities can be selected based on requirements, including 20, financial 2130, and market activity 2140. Icons 2142 are also provided for routine tasks such as email, calendaring, and document creation to facilitate quick navigation from one activity to another. A direct link 2146 is also provided to allow for the transfer of news or other items of interest. Various profiles can be selected based on where the user is. This is, for example, work,
At home or on vacation. This profile can be added 2170 when the user needs a new profile for another location. Various items of personal information 2180 are collected from the user to support various attempts.
In addition, a permission 2150 is set for each item 2180 to confirm that the information is current and current.

FIG. 22 illustrates a user main display according to a preferred embodiment. World news 2200 and local news 2210 based on user preferences
Is provided. The user has also selected real estate 2230 as an item that provides information directly to the main display. Further, different agents 2220 are provided based on user preferences.

FIG. 23 illustrates an agent interaction according to a preferred embodiment. Agent 2310 communicates information 2300 to the user, indicates that the user's life insurance needs have changed, and presents the user with a chart that best summarizes information for the user. Specific hints 2395 are provided to prompt for more detailed information based on current user statistics. A chart 2370 of the user's life insurance needs is also highlighted in the center of the display to assist the user in determining appropriate behavior. Buttons 2380 are provided to facilitate changing the policy, and a set of buttons 2390 are provided to assist in selecting various indications regarding the user's insurance requirements.

Event Backgrounder: The event backgrounder is a short description of the upcoming event sent to the user just before the event. The event background is constantly updated with the latest information related to this event. Relevant information such as itinerary and logistics is included, as well as other useful information such as acquaintances of users who may be at the same location. The purpose of the event backgrounder is to derive from a number of resources such as public websites and user calendars and contact lists.
Providing the most up-to-date information about an event and allowing the user to react appropriately in certain situations.

Vicinity Friend Finder This software allows a friend, family, or acquaintance to be in the same neighborhood as the user,
Or look for an opportunity to tell the user that you are coming to the same neighborhood. This software is
Scan the user's calendar for upcoming events. The terrain map is then used to compare these calendar events with those of the person listed in the contact. The user is then notified of any matches, thereby informing the user that there is a person scheduled to be near the user at a particular time.

Overload: The term overload has now become relatively understood, both in its definition and in meaning and consequences. Although the amount of attention available to humans at a time is limited, the amount of attention required is increasing every day. Simply put, too much information and too little time are the main factors that complicate the lives of knowledge workers today.

In a first attempt to dynamically address information overload, we focus on performing intelligent filtering of the information to reduce the amount of information. However, rather than simply removing small pieces of information randomly, most of these approaches
They are trying to be intelligent about what information will eventually be presented to the user.
This is achieved by evaluating each document based on the user's interest and discarding less relevant ones. Therefore, as a result, the quality is also improved.

[0139] Information filtering is only the first step in addressing information in this new era. Perhaps as important as the quality of the document is its easy access. Once you've attended a meeting, there is little value in getting a document containing important information about the subject of the meeting to your office. As the speed of business is spurred by the technology of intercommunication, the ability to receive high-quality information at any time and in any place becomes essential. This new approach, called intelligent information transmission, pioneered the new information itself.

The preferred embodiment demonstrates the theory of intelligent signaling described above in an attempt to not only reduce overload, but also deliver high quality information where and when the user seeks. In other words, the system communicates the right information to the right person, at the right time, at the right place.

Description of Active Knowledge Management System: FIG. 24 is a block diagram of an active knowledge management system according to a preferred embodiment. The system comprises the following parts: a backend 2400 connection to one or more servers, and a personal mobile wireless client (awareness machine) 2
430, 2436 and public client (magic wall) 2410,
2420, Web clients 2446 and 2448, and e-mail clients 2450 and 2460.

Pack End Server (2400) Process: FIG. 25 is a block diagram of a back end server according to a preferred embodiment. This back end (2400 in FIG. 24) includes the following software activity elements: an intelligent agent coordinator (Munin) 2580, an information priority determination subsystem 2530, and intelligent information collection and processing that is performed continuously and intermittently. Agents 2500, 2502, and 2504, user profile database 2542 and support software, information channel database 2542 and support software, and communication software 25
50, information conversion software 2560, and auxiliary software.

Awareness machines (2446 and 2448 in FIG. 24): Awareness machines are a combination of hardware devices and software applications. The hardware is composed of a handheld computer and a wireless communication device. Awareness machines reflect the constantly updated state of the owner's world by continuously receiving wireless information trickles. This information is mined and processed by a set of intelligent agents, including email messages, news tailored to each user's preferences, schedule updates, upcoming meetings and event background information, as well as weather information and information. Consists of traffic information.

The intelligent agent coordinator 2580 of FIG. 25 is also the user's “interface” to the system, and whenever a user interacts with the system, regardless of the GUI or other end-user interface, they will ultimately be intelligent agents. Interact with the coordinator (ask questions and send commands). The intelligent agent coordinator has four main responsibilities: 1) monitoring user activity, 2) processing information requests, 3) maintaining each user's profile, and 4).
This is to determine the route of information exchanged between the user and each agent.

Monitoring User Activity: Whenever a user activates a sensor, the intelligent agent coordinator receives an “environment queue”. This queue not only allows the intelligent agent coordinator to gain an understanding of where the user is for communication purposes, but also the standard patterns of each person's life (time of arrival,
(Departure time, etc.). These patterns are constantly being updated and refined in an attempt to increase the intelligence of the system when communicating information. For example, it is not uncommon for a single person to have several email accounts (work-based, home-based, mobile-based, etc.), and several different computers are involved in the search process for all such accounts. Is not uncommon. Therefore, in order for the intelligent agent coordinator to successfully transmit information to the right place, to maximize the probability that the user will see the information,
You need to consider all of these accounts and the times when users may access them. This is discussed further in another section.

Processing Information Requests: The intelligent agent coordinator personalizes the information with the intention of each user, and in order to more accurately reflect the interests of each user in the information they receive, from other agents. Handle information requests. Such requests are generally related to the user's profile. For example,
When an agent is creating a traffic report for a user, the agent may request the user's traffic area (search string) from the intelligent agent coordinator. All access to the user's profile data is accessed in this manner.

Maintain User Profile: The user profile contains extensive information about the user. This information
It is a mixture of user-specified data and information that the intelligent agent coordinator has learned and estimated from each user's information and activities. To protect the data contained in the profile, the intelligent agent coordinator needs to handle all information requests. The intelligent agent coordinator constantly modifies and updates these profiles by observing the user's activities and trying to learn their life patterns to support more routine daily tasks. The intelligent agent coordinator also utilizes other agents to pick up meaning from each user's daily activities. These agents try to dig up this data and find indicators of current interests, long-term interests, and even time preferences for communicating each type of information. Another important aspect of intelligent agent coordinator observation is trying to determine where each user is physically located throughout the day for routing purposes.

Information Routing: Most people move throughout the day. The intelligent agent coordinator, through both observation (unsupervised learning) and queues from the environment,
Try to be sensitive to this fact by trying to determine where the user is, or likely to be. This is definitely important not only to determine where to send the user's information, but also to determine in what format to send the information. For example, if the user is at a desk and is using a web client, the intelligent agent coordinator should receive an indication of activity from the user's PC and send any necessary information to it. I understand. In addition, desktop PCs are usually so powerful that they can send a full-featured version of the image. However, there may be an alternative situation where the Intelligent Agent Coordinator receives (by a key card reader near the exit) an indication that you have just left the building. A few minutes later, the Intelligent Agent Coordinator receives notification that you have received an urgent message for you. The Intelligent Agent Coordinator knows you have left the building and has not received any other activity
Consider the handheld device (also knows its capabilities) to be reachable and send the text of the emergency message there instead of the image-first version.

Inherited Innovation: Active knowledge management systems represent some of the most sophisticated thoughts in the world of knowledge management and human and computer interaction. Some of the key innovations include: ● The intelligent agent coordinator illustrated above. ● Development, demonstration, and realization of the theory of intelligent information transmission. ● Support for several channels of communication, all utilizing a common backend. For example, if the user is in front of a magic wall, the information is presented in a multimedia-enabled format. If the system determines that the user is traveling,
The information is sent to the awareness machine in standard text. This facilitates the transmission of information regardless of the time and place where the user seeks the information. ● Personalization of information not only based on a static user profile, but also by taking into account the user's interaction history and current real-time situation, including awareness of "people, places and times". Use of a fast and scalable information priority decision subsystem that takes into account the intelligence of the intelligent agent coordinator, user preferences, and user interaction history. By removing the burden of everyday judgment from the intelligent agent part, the agent can be made more sophisticated and accurate without compromising the scalability of the system. ● Offers efficient, intuitive, and emotionally satisfying interaction with systems through speech recognition and speech synthesis combined with intelligent agent animation and tactile input.

Client Reporting Subsystem Model: Context: The reporting subsystem is used by other subsystems on the client to perform reports (read, create records). This subsystem does not assume the type of data it will handle-this data could be erroneous reports (as part of an architectural service) or lead management information (as part of an application data service). The reporting subsystem, in this sense, is part of the infrastructure and is a set of basic services available everywhere on the client. The reporting subsystem uses the communication subsystem to store and transfer data.

Architectural overview: The reporting subsystem provides services to any client subsystem. This subsystem provides a mechanism for passing messages within the client application and between the client and server. This reporting mechanism uses the communication subsystem to store and transfer data. Exceptions use this reporting mechanism to report only information about errors.

Role: The reporting mechanism provides a set of infrastructure services that allows architectural components to report information. This subsystem does not specify what information is reported, but includes components that map to specific types of report information, such as system faults or customer interaction information.

Some of the components of this subsystem are presented as a set of exceptions,
This allows for automatic reporting of error conditions encountered during processing.

This subsystem accepts the data and transfers it to the communication subsystem in the appropriate format. This subsystem catches and reports exceptions generated during processing that are the result of an error condition. This subsystem can handle any type of report that needs to be created, from error logging to sales leads. It is flexible enough to record new types of information as needed. In addition, it is flexible enough to add new types of reports as needed. In addition, situations where certain information is not available during the logging process can be addressed.

Responsibility: The client reporting subsystem is responsible for providing a set of exceptions for use in every part of the application. This subsystem is further responsible for logging failure reports, application heartbeat reports, message receipts, referrals or leads, and management information system entries.

Disclaimer: This subsystem is not responsible for gathering information from interactions at the interface or elsewhere, nor is it responsible for determining which datasets need to be reported. I do not have. Reporting does not include printing the report.

Component Specification: Client Exception Reporting Component This is a set of exception classes,
Use the reporting service of the client reporting component to store and send fault information.

Client Reporting Component This is the mechanism that collects, formats, and sends information for any reports to the client communication subsystem.

Creation, persistence, and management: A key element of this subsystem is the static class-report factory that manages the creation of report objects. This class is instantiated by the communication subsystem (which manages the configuration of the client) and is always available. Otherwise, it is a serious error. Reports are generated on an ad hoc basis as needed.

Sizing and Capacity: Whatever the requirements of the client architecture, the “throughput” of the client reporting subsystem (understanding the number of reports required at a given time, the total number of types) Do not overburden resources.
Most of the capacity requirements for reports are absorbed by the communications subsystem that needs to be prepared for storage and transmission of such reports.

Nevertheless, this subsystem must be able to handle any throughput required by the architecture, and its design takes into account the estimated work generated by each part of the architecture.

Performance: Performance is not important in the client reporting subsystem. Report data is
With a few exceptions, the delay before the data is transmitted is expected because it is stored before transmission. Certain types of serious or critical failures need to be reported at once, but the low bandwidth required for such transmissions does not present a performance problem.

Design Guidelines: While architectural reporting needs fluctuate, a core set of capabilities (failure reports, lead management, interaction reports) always remain a requirement. This subsystem has the flexibility to allow not only expansion or reduction of capabilities but also adjustment of the detail and nature of the data recorded for each capability. Implementation of this system is in accordance with the Project Java coding standard .

Logic component: An exception in the client exception reporting component calls the client reporting component and creates a fault report.

Component Description: Client Exception Component This is a hierarchy of exception classes, built to support exception handling and passing. These exceptions accept information about the exception event that they represent. With this information and everything else that this class knows about its events, this component uses the client reporting component to create failure reports.

Client Reporting Component Static report factories accept requests from other components in the form of signaled events. Based on this event, the factory creates a specific type of report. This report is then populated with information provided by the calling component or reporter.

Publish Report: Component A signals the client reporting subsystem to indicate that an event to report has occurred. Next, the client reporting subsystem requests information about the event and creates a report. Finally, the client reporting subsystem signals to the client communication subsystem that it is ready to send a report.

Throwing an Exception: First, Component A signals that an exceptional event has occurred by creating an instance of the appropriate exception. Second, component A is
Pass the relevant information of the event to be reported to the exception. Third, component A requires that an exception be thrown. Fourth, the client exception reporting component emits a report to be sent based on the available information using the client reporting component outlined above. Finally, the client exception reporting component throws an exception.

Local Content Subsystem Model: Context: The local content subsystem provides any content required by the application. This includes both static and dynamic content. In addition, it provides business services required by applications.
It operates based on "storage and retrieval", stores data obtained from a user or a company, and provides a mechanism for retrieving the data. The local content subsystem is used by the ISF subsystem to provide content. Receive business data using the communications subsystem.

Architecture Overview: The objects of the local content subsystem are created by the initialization subsystem of the application architecture. Services provided by the local content subsystem are further accessed through the application architecture through the initialization subsystem and the ISF subsystem.

Examples of content in the local content subsystem reflect various types of business knowledge and processes required by the application. User data and business data involve data collected from users and businesses, respectively. The calculation component performs complex calculations, and the product component represents the product used by the enterprise. The content provider component defines static media content.

Role: The local content subsystem provides static and dynamic content to the application. In addition, it provides any business-specific services required by the application.

Responsibilities: The local content subsystem provides static and dynamic media for applications. The local content subsystem also stores details entered by the user, as well as business data, and performs calculations regarding the business.

Disclaimer: All application-specific operations are provided through application personalities. This behavior is defined by Hamlet scripts, which also define the navigation between scripts. This script is divided into metaphors, each of which is an interactive style embodiment.

Access to media is provided through content provider components that belong to this subsystem. However, the object for this component is derived from the content file defined in the application personality.
Created automatically by the system initializer component.

Creation, Persistence, and Management: Business data components are created and initialized at the time of system initialization. This persists for the life of the system.

[0177] The user data component is available throughout the time that the customer is using the system. At the end of the customer's session, references to all objects in the user data component are released so that the objects can be garbage collected.

Objects in the local content subsystem (except for the business data component) are created and initialized internally or by the initialization subsystem. References to these objects are managed by the ISF subsystem. Business data objects are created, initialized, and managed by the communications subsystem.

Logical components: The ISF has no actual knowledge of the local content subsystem. The local content subsystem implements a set of interfaces defined by the ISF. The initialization subsystem uses scripts defined in the application personality to define which objects (implementing these interfaces) need to be used to search for content. The ISF uses an initialization component to create these objects and then manage them.

A developer of an application personality can easily grasp his or her own script by directly managing the local content object. This allows local content objects to be developed without knowledge of the application architecture layer.

Component Description: User Data Stores and retrieves data entered by the user and initiates calculations on the stored data. Business Data Store and retrieve data provided by the company. Calculation Perform complex calculations. Products Provide access to information related to specific marketing products. Content Provider Provides access to static media. This is an implementation of the interface and is not documented as a separate component.

Interface Support Framework Subsystem Model: Context: The interface support framework subsystem is part of the application architecture layer. Provides a rich interactive environment that takes advantage of all the possibilities of a dynamic multimedia interface. This ISF is
It is built around a theater metaphor where all objects are expected to perform dynamic actions.

Objects in the ISF are initialized by the application initialization subsystem in the application architecture layer and utilize the content player, printing and reporting subsystems of the technical architecture layer.

Architecture Overview: Objects in the ISF are initialized by the application initialization subsystem. The reporting and trading interface service is used to log, report or print ISF data for the technical architecture layer. The content player subsystem within the technical architecture is used by the ISF to present media to users.

The ISF subsystem is built on a layered architecture that follows the model-view-controller pattern. The fact components include the business object model and the definition of the business media content. The visual component displays and manipulates media to provide a user with a view of the business model. The behavior component controls any interaction between the visual component and the fact component.

Role: The ISF subsystem provides services to applications to present multimedia content in a controlled manner. In addition, it provides the ability to respond to user input and affect scene changes.

Responsibility: The ISF subsystem displays each scene of the application and modifies the content of the scene while it is being displayed. In addition, the ISF subsystem enables navigation between scenes, responds to user interaction, retrieves business content, performs business functions and calculations, and provides a common user interface construct. Other responsibilities of the ISF include starting print jobs and video conferencing sessions, reporting user entries to scenes, session duration, user interaction with roles, user navigation, ISF
Reporting errors that have occurred within. Finally, the ISF manages user sessions and responds to system level events such as system startup and shutdown or screen paint requests.

Design Guidelines: The ISF subsystem services applications to present multimedia content. Architecturally, it is hierarchized into three separate components corresponding to the model-view-controller paradigm. This separates the core business object and its data (model) from the visual representation (view) of this information, and the logic (controller) that controls the model or view and reacts to these changes. This architecture provides the boundaries between the system's graphical style (stages, roles, and scenes), the operating code (actors, scene directors, and stage managers), and the underlying content providers (business objects). . These sections are the visual, operational, and factual components.

The fact layer does not recognize the visual layer. This allows the visual metaphor to change without disrupting the underlying business domain model. The motion level mediates between the fact and visual layers, and very complex interactions with either layer should be avoided. Where possible, anonymous communication with a publish / subscribe pattern is used to avoid further interdependencies between layers.

A stage is specified as a display context. It can only communicate with the controlling location. This is hidden behind the stage manager, who needs to manage all visual requirements.

[0191] ISF is a hierarchical system. All roles in the scene form a series of visual siblings. These roles actually encapsulate other roles. This allows any number of distinct processing layers through recursion. Each child communicates only with its immediate parent and relinquishes control of communication beyond that parent. This containment relationship is possible in both the visual and motion layers.

To support navigation, scene thumbnails are maintained. The user can return to a previously visited scene by touching the scene thumbnail.

Component Description: Visual Component User interaction and presentation of multimedia content. Operational components Search application operations and multimedia contents. Fact component Provides multimedia content and business function services and calculations.

Process Control: This table describes the various key threads that run within the ISF.

[0195]

[Table 6]

User Touches Location on Stage: Description: The end user touches the stage, the visible window of the application. As a result, a response from the application is started. Actor: End user Components involved: Visual Key objects involved: Stage

Stage Handles User Touch: Description: The application window determines which location is affected by the area touched and notifies the corresponding touch role. This stage also controls the visual cues displayed on the window. Actor: End user Components involved: Visual Key objects involved: Timer, User Interaction Reporter, Location, Media Player

Roles Accept User Touch Events: Description: Each role is notified of a user touch. This forces the role to request a change of the medium in the corresponding medium and, upon completion, notifies the actor of the user interaction. Actor: Stage Components involved: Visual, Behavior Key objects involved: Actor, Role

Actor Activates Event Casting: Description: Actors cycle through all registered casting lists and activate all castings that are interested in a particular event. Casting behaves in a polymorphic manner, so the behavior of how it reacts is held in the casting, not the actor. Actor: Role, Execution Casting Components involved; Behavior Key Objects involved: Actors, Casting, Stage Manager,
Scene director

Stage Manager Performs Scene Transition: Description: The stage manager replaces the currently active scene with a new scene based on the information in the navigation casting. It is important to control how the change occurs and maintain the visual illusion of the kiosk world.

This scenario is also called when the application is started or restarted.
In this case, there is no current scene, but the application is commanded to transition to the first scene of the application.

Actors: Navigation Casting, System Initializer Components Involved: Motion, Vision Key Objects Involved: Stage, Scene Director, Session Manager

Scene Director Performs Cast Change: Description: The scene director coordinates the activation of all timings so that any messages to the stage manager are grouped together and coordinated at the appropriate time. This allows any changes in roles that can be seen on the scene to occur simultaneously.

Actors: Content Casting, Slide Casting Components Involved: Vision, Behavior Key Objects Involved: Scene Director, Stage Manager

Scene Director Performs Cast Change With Sliding Effect: Description: The scene director raises all non-sliding roles to the stage, and then displays the slide casting content along a straight line to its final destination There is a need to.

Actor: slide casting Components involved: vision, movement

Calling a Business Function with a Business Object: Description: An application can request business information from a business domain model (eg, return a payment). The interface also supports entering values into the business object (eg, storing it as a loan term amount). Each business object is provided with a generic interface for invoking actions. The desired behavior is specified in the business function casting, which can return information to the actor associated with the business object.

Actor: Business Function Casting Components Involved: Facts

Stage Manager Timeout from Inactive State: The application needs to support a timeout function in case the user leaves the application before returning to the attractor scene. This protects the privacy of the details input by the user. Actor: Wait timer Components involved: Operation

Session Manager Reset Application: Description: During the execution of the application, it may be necessary to reset the stage to the first scene and clear the session. This can be triggered by the inactivity of the system or by a direct user request dispatched through a business object.

Actor: stage manager wait timer, business object Components involved: operation

Reset session information: Each session stores information collected about the user. At the end of the session or at the request of the user, all input data can be deleted. This supports privacy.

Actor: Session Manager Components involved: Behavior

Reaction Medium Displays Medium: Description: The visual component of the ISF is responsible for displaying any medium (image, audio, video, text) on the stage. It actually interfaces with the underlying media subsystem in the client technical architecture.
Each player supports all the code needed to present the media to the user, obtained through the gatekeeper.

Actors: Roles Components Involved: Vision, Behavior, Facts, Content Player Key Objects Involved: Reaction Medium, Medium Player, Stage

Application Requests Hardcopy Printing: Description: The client application can use either static information (checklist) or dynamic information (current interest rate and product, product simulation, and / or other dynamic components of a line graph). And printing of product description). This is initiated by the end user through print casting.

Actor: Print Casting Components Involved: Operation, Printing Key Objects Involved: Business Object Reporting Interface Subsystem Model

Context: The reporting interface subsystem collects information logged by the application architecture layer and sends it to the technical architecture layer client reporting subsystem.

Architectural Overview: Reporting interface information is logged by components in the application architecture layer and sent to the technical architecture layer client reporting subsystem.

Role: The reporting interface subsystem provides a service that logs user interactions with the kiosk and reports software and hardware faults that have occurred within the application architecture layer.

Responsibilities: The reporting interface subsystem collects user interactions with the kiosk by capturing what the user has done to the system, eg, what scenes the user is visiting and what visual elements they are interacting with. And responsible for logging. The reporting interface subsystem also collects and logs information about the business product that the customer is interested in, the output of the business function that the customer invoked, and the business data that the customer entered. Finally, the reporting interface subsystem captures information about the kiosk's software and hardware performance. This information can then be used for error handling and fault management analysis.

Disclaimer: The reporting interface subsystem does not include services to collect and log customer interaction information such as customer names and phone numbers.

System Management Subsystem Model: Context: System management involves the definition of a combination of automatic and manual procedures. Automation is achieved primarily through the use of a System Management Server (SMS). S
MS is a tool in the Microsoft Backoffce tool suite, and can centrally manage system software and hardware in a distributed environment.

Architecture of the System Management Subsystem Model: The architecture of the system management subsystem model consists of two components, Syst
It consists of em Management Server (SMS) and fault monitoring.

System Management Server (SMS): The System Management Server (SMS) is a Microsoft tool that can be used to distribute software / content, perform software audits, perform fault diagnosis, and perform remote control. . SMS is supplemented by a component developed for this architecture, the file transfer utility.

Fault Monitoring: Kiosks are monitored in real time through a heartbeat message system. Heartbeat pulses are sent from the kiosk at a set rate (eg, every minute) and are monitored by a console running a kiosk monitoring application. If the status of the kiosk changes to indicate a failure, the monitoring application initiates the appropriate action. Some errors are handled by existing operations centers. The routing of such errors is covered by the application services subsystem.

Role: SMS is used for medium-scale software distribution (both code and content) and fault diagnosis of remote kiosks. There are a number of features that make SMS flexible and effective support functions. Fault diagnosis provides a means to view the real-time status of each kiosk and its associated peripherals. When a problem (out of paper, etc.) occurs at a particular kiosk, the kiosk is brought to the attention of operations personnel. You can observe the status of the kiosk to verify the solution to the problem.

Responsibilities: SMS is responsible for software distribution, hardware fault management and diagnosis, and remote reboot of clients. SMS also provides a user interface for selectively viewing the status of any kiosk in the network.

Creation, Persistence, and Management: SMS resides on a dedicated server according to the preferred embodiment and is always available.

Performance: The elapsed time between the occurrence of a failure on the MMT and the subsequent display of the failure on the operation terminal depends on the frequency of the heartbeat, the ability of the application server to process the heartbeat, and the operation terminal. Changes according to the frequency of refreshing.

An example according to a preferred embodiment is provided below.

[0232]

[Table 7]

If no heartbeat message is received from the kiosk within 5 minutes (configurable), fault monitoring sets the kiosk status to unknown. This time lag is
This is necessary to avoid falsely reporting the MMT machine as unknown when the actual problem exists slightly later than the normal processing time of the heartbeat message.

Logical components: An SMS site is composed of two components-a primary site and a client. The primary site is the top level of the SMS hierarchy. This site contains its own SQL database for storing system and inventory information about itself and the secondary sites below. Clients (kiosks) are managed from a primary site. The client sends hardware / software information to the SMS server through the SMS inventory service.

Fault Monitoring: After a configurable interval, the client checks the status of the machine, printer, and application and sends the result to the server in a heartbeat message. The server then puts this status into a kiosk status database where fault monitoring is performed by operations staff.

Server Communication Subsystem Model: Context: The server communication subsystem model is part of the server technical architecture.

The server communication subsystem handles all communications between the client, the central server, and the mainframe host.

Architecture Overview: The Asynchronous Messaging component provides asynchronous message-based communication between the client and server using standard Internet mail protocols, SMTP, and POP3.

[0239] The business process access module component provides a common point for invoking predefined business functions, such as recording interaction information from the MMT. HTTP is a protocol used to communicate with an HTTP server on the World Wide Web. This is used by the MMT to distribute small updates of application components and content during the startup client configuration process. Access to the mainframe resident database is performed by duplicating the required database tables into the corresponding server resident database tables. The reverse process is used to centrally store data stored on the server in tables in the mainframe database. General alerts from the server are transmitted to the mainframe through an interface with the mainframe front-end processor.

Role: The role of the server communication subsystem is to provide the server with MMT (or Internet).
It is to provide a communication function between a client and a network server and between a network server and a mainframe system. In addition, the server communication subsystem isolates and provides access to organization-specific functions.

Responsibilities: The server communication subsystem follows standard Internet protocols and allows easy porting to its transmission channels. In addition, the server communication subsystem provides reliable asynchronous communication between the client and the application server, and controlled and reliable access to organization-specific functions. Finally, the server communication subsystem provides the ability to communicate updates to client settings, such as application components and content, and provides the organization's legacy through predefined processes such as database replication and general alert reporting. Provide access to the system.

Disclaimer: Server communication is limited to calling modules that comply with the ACT messaging architecture. If communication with another platform is required, it needs to be located in an external system module that uses the organization's messaging or access method.

Transaction Interface Subsystem Model: Architecture Overview: Transactions are initiated by components in the application architecture layer and sent to the client reporting subsystem in the technical architecture layer.

The transaction services specified in this document are not implemented as stand-alone components by themselves, but are implemented as extensions of existing application architecture components.

Role: The transaction interface subsystem is responsible for providing an interface to the application to store content information about the end user. This includes, in part, the information needed to complete the loan, survey-based information on customer demographics, account balance questions, and fund transfer.

Customer Lead Transaction Execution: The Customer Lead Transaction Execution application facilitates the interface support framework and enables the services of the technical architecture reporting subsystem to support the collection and storage of information about end users.

Disclaimer: The current usage of the trading interface assumes that only asynchronous communication is available.

Server Application Service Subsystem Model: Architecture Overview: This application service subsystem contains a set of definitions for building MMT server applications and architecture modules.

Role: The server application service subsystem contains a set of services and definitions for accessing application and architectural functions. This subsystem defines the structure and support services for building and executing applications and modules on an application server or operational workstation terminal. Supported features include transaction processing with the MMT, for example,
It includes customer introduction information, customer interaction information, MIS information, fault information, product rate and price information, application setting information, message receipt information, and heartbeat status information.

Responsibilities: The server application service subsystem handles application business logic on the server independently of the underlying database management system. The application business logic includes customer introduction information, customer interaction information, MIS information, fault information, product rate and price information, application setting information, message receipt information, and heartbeat status information.

The server application service subsystem also logs errors, decodes certain codes (eg, “1” = NSW, “2” = QLD, etc.), and retrieves configuration information from the registry located on each machine. Provide common services that server and client applications can use to perform searches

Finally, the server application service subsystem invokes a business process (BP) for certain BP messages.

Logical components: The server application service subsystem includes common services, data access modules, business processes, and business process access modules. The definition of each component is as follows.

Component Description: Definition: Common Service. Decode the code table, retrieve the setting information from the registry,
A common service that supports the development of application functions, including message processing and support for error logging and processing of server applications.

Data access module. A data access module (DAM) provides access to data in an application database. DAM is inserted,
Perform specific data accesses, such as delete, update, select, or select all of one or more tables. This is an MFC extension DLL that encapsulates a recordset object that uses ODBC to access the underlying DBMS. The DAM definition outlines how such modules are used and coded.

Business process. A business process (BP) is an application or architectural function that can be invoked by a business process access module architecture. The BP is specified by the message type. BP
Can accept the request message defined by the BP and provide a response message to the synchronization message. Database access is provided to BP by DAMS. BP is MFC extended DL with defined entry points
L.

Business Process Access Module. This component is described in detail in Communication Subsystem. Business Process Access Module (BPAM)
Is an architectural component that provides access to business processes. BPAM invokes BP for certain message types. The BPAM accesses the message address table and looks up the details of the BP module. This component is described in detail in Communication Subsystem.

Wireless Electronic Bullet According to Preferred Embodiment: One embodiment of the present invention is a mobile portal platform that includes a mobile portal and an electronic bullet. An electronic bullet is a handheld wireless computer device that runs thin client software. Electronic Bullet
Various sensors such as GPS, biosensors, and environmental sensors are integrated. In addition, recording devices such as cameras and audio recorders are also integrated into the electronic bullet. Mobile portals include mobile portal servers that are connected to various third-party content and service providers through the Internet or mobile portal extranet.

FIG. 26 is a flowchart showing how the hardware and software of one embodiment of the present invention operate. The electronic bullet 2602 receives input data from sensors, GPS, cameras, microphones, and other user inputs 2600 integrated into the wireless handheld device. Electronic bullet 26
The thin client application running on O.02 allows the electronic bullet 2602 to execute many different software applications without requiring a large amount of internal memory and storage capacity, as described in more detail below.
Electronic bullet 2602 forms a message based on the received data and user input. Thereafter, the electronic bullet 2602 transmits the message to the mobile portal 2606 via the antenna 2604. The mobile portal 2606 parses the message received from the electronic bullet 2602 and forms a new message based on the received message. Next, the mobile portal 2606 determines an appropriate third party service provider 2608 to transmit the new message based on the content of the message received from the wireless handheld device 2602, and then transmits the new message. Next, the third party service provider performs the appropriate service and returns the results to the mobile portal 2606. Thereafter, the mobile portal forms a message based on the data received from the third party service provider 2608 and sends the message back to the electronic bulletin 2602. Thereafter, the electronic bulletin formats and displays the received data. Electronic Bullet 2602
A wireless modem such as an OCET SE Wireless Modem is used.

Of course, wireless performance is not as reliable as a conventional dial-up connection. In some places in San Francisco, you could get a strong connection as long as you were near the window. However, at CNET's total brick headquarters, Ricoc
het could not connect at all. When connected online, performance up to 28.8 kbps is available with a slow drop to slower speeds. However, there was no disappointment at slow speeds. Alternative method-connection by cellular modem-
Ricochet is much faster, more reliable, and has lower usage costs.
Of course, this SE Wireless is battery powered. The continuous battery life of this modem is up to 12 hours.

Therefore, the user can use the mobile portal 2606 via the electronic bullet 2602 by using the wireless modem. Using the appropriate key (s), the user can select a service to use in response to appropriate data obtained from sensors, GPS, cameras, microphones, and other user inputs 2600. .

In certain circumstances, data can be automatically sent to a selected service based on the type and value of data obtained from electronic bullet 2602. For example, when the integrated biosensor obtains a particular predefined data value, the appropriate emergency care provider is automatically contacted. In addition, data obtained from sensors, GPS, cameras, microphones, and other user inputs 2600 may be combined before sending to the appropriate service provider. For example, in the above example, GPS location data may be sent to the emergency care provider along with biosensor data. The emergency care provider will then know the patient's biological data and the patient's location. Thereafter, appropriate services are provided.

Mobile Portal Platform: The mobile portal platform is a high-impact server-based application according to the preferred embodiment, which services each customer through a personal site located on the server and provides a personalized experience. It focuses on the theme of doing. Intuitively organized, this service is intended for customers' intentions-basic living needs or decisions that require extensive planning, and some issues such as financial planning, health care, personal and professional development, family life, and other issues. It is intuitively organized with a focus on satisfying adjustments across the dimensions. Each member has and maintains their own profile, which allows them to create and browse content in systems that target themselves in a limited manner. From the point of input of the service request, research is performed, transactions are executed, and advice is provided using the intelligent agent. By using advanced profiling and filtering, intelligent agents learn about users and improve the services they provide.

In a preferred embodiment of the system, Windows CE with GPS receiver
Use PDA. This embodiment is configured for a shopping center including a plurality of stores. The system uses a GPS receiver to determine the location of the user. One of the advantages of this system is that it allows data to be searched at nearby stores without relying on the presence of any special equipment at the shopping center itself. The accuracy of a small and inexpensive receiver is approximately 75 to 100 feet (2
3 to 31 m), which has so far proven to be all that is needed to accurately identify nearby stores. This system is
Use generated data instead of actual store ads and prices. A well structured online catalog is used. Other embodiments utilize a "learn to stop" agent at certain stores using a relatively small amount of knowledge. Furthermore, as retailers begin to use standard packages for using online catalogs, the number of different formats is expected to decrease, and as a result,
The number of competing formats will be easier to handle. As e-commerce evolves, it is not unreasonable to expect standards that will evolve the management of how product offerings are represented.

Purpose designation: Before going shopping, the shopper creates a list of items to purchase from an existing set of about 85 product categories (men's casual pants, women's formal shoes, flowers, etc.). The shopper also indicates from the list of shopping centers the shopping location he is going to visit.

Initial store selection: After arriving at the shopping center, the nearest store that initially sells at least one item of the type entered by the user during purpose specification. Along with the store name,
The system of the preferred embodiment creates a list of specific items available for purchase and their prices. The map of the shopping center displays both the exact location of the store and the current location of the shopper. The shopper inquires at any time to indicate a store based on the current location.

Browsing: Addresses the needs of many shoppers visiting or shopping at a shopping center, generally without deciding on a particular destination. FIG. 27A illustrates a display according to a preferred embodiment of the present invention. The display operates in a browse mode used by shoppers when strolling through a shopping center. In the browse mode, the system indicates items of interest that are currently being sold at the store closest to the shopper. An item is considered of interest if it matches the category entered on the destination screen. If there is no item of interest, the type of common product sold at that store is displayed instead of the specific item. As the shopper walks, the map displays the exact current location of the shopper within the shopping center. If the shopper selects the item displayed during browsing, the system alerts the shopper to a local retailer offering the same product at a lower price, or informs the shopper of the local lowest price. . This search is limited to local shopping malls, which are areas where shoppers are assumed to have the will to travel.

Alternative: It is worth emphasizing that the agent of the present invention supports a wide range of shopping tasks and can operate, for example, as an interactive channel. That is,
Not only can the information be provided to the shopper, but also to the retailer at the shopper's discretion. In this embodiment, the system indicates the shopper's purpose and preferences to the retailer-based agent, who then responds with a customized offer that bundles services with the product. Allowing customization of the offer is very important for obtaining the cooperation of retailers who do not want to compete on price alone, and is valuable for customers who purchase on a basis other than price. Although the preferred embodiment focuses primarily on location-based filtering in the context of a shopping task, the present invention provides a "physical task support" that provides an information channel to those engaged in various tasks in the physical world. Provides the basis for agents.

Location Predictions: The present invention has significant advantages over agents that are not location-based, since the user's physical location is often very effective in predicting the user's task. If you find that someone is at a bowling alley or post office, you can reasonably estimate your current activities. Knowledge of a user's current task largely determines the type of information that the user is likely to find useful. It is unlikely that a bowler will pay the postage, or a stamp buyer will worry about the optimal bowling ball weight. In addition, knowledge of resources and obstacles at a particular location indicates a range of actions that are possible and likely to be performed by a person at that location. This perception of actions that are possible and likely to be performed by the user can be used to further restrict the types of information that the user is likely to find useful.
For example, knowledge of a restaurant's wine list can be used by a recommendation system to constrain the wine advice presented.

Knowledge of the exact location of the shopper at the shopping center is valuable because it allows the localization of shops near the shopper. The offer content of the store closest to the shopper means that the shopper can immediately select. If the shopper focuses on looking directly at the product and there is a cost associated with going to another store, then the product at the nearest nearby store is the most likely immediate option for the shopper . As a result, the most valuable information provided at any given time is the availability of merchandise in surrounding stores that is consistent with the previously stated objectives.

Humans tend to move to different places while performing their many tasks.
This indicates that the person's vicinity has not fully seized the full range of options available. In fact, one of the major reasons for leaving a location is to perform actions that are not possible at the current location. Nevertheless, humans tend to cope with most tasks in relatively close areas. Thus, one's vicinity often shows options available at a given point in time, and the perspective of a wider location often captures options that are likely to be considered in the course of performing a task. In the case of shopping at a shopping center, for example, stores near shoppers represent available options at that time. However, shoppers at the shopping center generally have the intent to move to any store within the shopping center. Thus, possible options for the entire shopping include all stores in the shopping. Therefore, the reason for presenting the information on the offering of interest only from nearby stores is that these are immediately available options. When asked for an alternative, the system limits the system itself to all stores within the shopping center-an area where the entire shopping task is likely to be performed. Shoppers need to look directly at the product in question, even though they are warned that a store hundreds or thousands of miles away is selling the same product for a few dollars less than the cheapest local alternative Or little value if you do not want to wait for shipment.

Physical Shopping vs. Online Shopping: In addition to the significant benefits over non-location based agents, the present invention overcomes the shortcomings of online (or Web) shopping. It is tempting to assert that online shopping will eventually become the dominant shopping form, with the further penetration of home computers, the expansion of online offerings, and only good online shopping tools. So, at first glance, it seems like a mistake to start using places to support activities that are later virtualized. Already, numerous software agents that support online shopping have emerged. For example, a program has been given a name that allows a user to specify the cheapest source of a music CD. A similar program is
It has also been developed for purchasing books, such as BargainBot. Such a system demonstrates the potential of e-commerce web agents to create a complete market for a particular product. The success of such agents will drive the development of similar Web shopping agents for a wider variety of products.

The Limits of Online Shopping: Undoubtedly, online shopping continues to grow and the trend toward more powerful online shopping agents continues. Nevertheless, Web
No matter how sophisticated agents are, it seems clear that traditional physical shopping will continue to dominate the market in the foreseeable future. Several difficulties inherent in online shopping ensure continued trust in physical shopping.

Insubstitutable: Web-based shopping agents typically allow users to specify the cheapest price for replaceable products such as books and music CDs. While the ability to create a “perfect market” for these products is a great benefit to consumers, there are several drawbacks that complicate applying such an approach to any product.

The merchandise is particularly suitable for shopping agents because it is easy to compare with competing offers. Since goods are interchangeable, very few dimensions that differ are prices. As such, price is the primary, if not the sole, criterion for making purchasing decisions.

However, as soon as we think beyond merchandise, several other criteria become important. For example, how do you compare items such as sweaters, mattresses, or tables? In addition to the price, pay attention to the material used, the color, the fit and comfort and the workmanship. Similar issues apply to most other products.

Inaccurate goal specification: Second, a related difficulty is in communicating your needs to agents. Shopping agents work great when users know exactly what they want. In this case, the user simply inputs the name of the product. Unfortunately, if you don't come up with a particular item when shopping, the problem of telling the agent what you want is more difficult. For example, how can a user tell an agent what kind of lamp they want to place in their living room?

Unclear Preference: An interface that allows shoppers to include descriptions of features such as price ranges, colors, options, brands, etc., helps to address the above issues,
This is not enough. In most cases, shoppers have not established preferences or cannot accurately represent what they want until they start shopping and have the opportunity to explore various examples of products of interest.

Shopping is Entertainment: Humans like shopping, and they shop even if they have no specific purchases in mind. In one study, 42% of consumers are "purposeless shoppers" who visit shopping centers for window shopping and social activities as leisure.

Shopping is intuitive: Even though the user can effectively provide these details, it is unlikely that most users will rely on these agents to make purchase decisions. After all, many people, even their trusted spouses, are anxious about making the right purchases for themselves. Most people want to see and touch directly what they are considering before making a purchase decision. The slight taste that can be offered to agents cannot be replaced with this rich, direct experience. These preferences are such that they can be used to generate a set of candidates for the shopper to consider.

Instant Satisfaction: Shopping is often a very emotional activity. Humans are often happy with their purchases and can't wait to go home and try. The intrinsic delay between online purchase and receipt is a major problem for those who need to take home immediately after confirmation.

In the end, consumers will continue to engage in physical shopping due to the limitations listed above. However, the fact that tasks cannot be completely delegated to software agents does not preclude the role of such agents. First, when knowing what they want, the agent is helpful to the user when it comes to purchasing goods. However, the second role is to support itself throughout the time a human is engaged in physical shopping tasks. This is, of course, the approach used in the Shoppers Eye Project.

Shoppers Eye: At first glance, the invention seems to be affected by some of the same limitations as pure Web-based agents. After all, is communicating your purpose to the PDA easier than communicating to a web-based agent? Will the purchases supported by the PDA system make my preferences clearer than with Web-based agents?

The main difference between a pure web-based agent and the current “physical task support agent” (ie, an agent that supports users engaging in tasks in physical situations) is that the web-based agent It is completely responsible for conveying any information considered by the user. On the other hand, the “physical task support” agent according to the preferred embodiment can augment the web-based agent approach by referring to aspects of the user's environment. For example, if there is a particular sweater at a store 30 feet (9 m) away, it is not extremely important to fully communicate the comfort of that sweater. All that is needed is to introduce the sweater to shoppers. The shopper can try on the sweater and get a better perception of the sweater than any information that can be conveyed by the system. If there are too many products matching the purpose incorrectly specified in the web-based agent, a more limited search needs to be performed. However, many agreements simply indicate that there are stores that are likely to be of great interest to shoppers and should be visited. Once in the store, it is often much easier to narrow the range of products that the person is interested in than to refine the purpose on a Web-based agent. Therefore, the physical task support agent is a means of communicating information throughout the task process by alerting the user to aspects of the physical environment, allowing the user to express his preferences and Can help identify the

Physical Shopping Agent Promise: It is not surprising that physical shopping has been ignored by the agent community. After all, until very recently, there was no reliable way to deliver personalized information to individual shoppers at a distance, however, the explosive development of PDAs and more sophisticated Communication capability promises to be an effective channel to deliver information "just in time" regardless of where the user is. The present invention leverages the promise of developing channels that support every stage of the shopping task, provides an intuitive new agent that supports physical shopping, and solves the aforementioned problems, including:

Specifying the Purpose: The shopper first indicates at least the general category of the item of interest. Shopping agents need to be able to specify the purpose in varying degrees of specificity. In the present invention, these objectives can be refined as the task progresses.

Exploring Product Space: Shoppers need to know what is available before they can make a choice. Shopping agents can assist with this task by presenting various offering classes, reviews, demonstrations, and more. The physical shopping agent of the present invention can reinforce this by providing shoppers with tours of available content available locally.

Elaboration of preferences: As shoppers learn what is available and scrutinize offerings, shoppers' preferences evolve. Agents need to enable shoppers to refine preferences over time. The present invention allows a user to refine preferences.

Identification and Comparison of Candidate Products: As shoppers begin to understand what they want and what is available, they usually compile a list of candidates to consider more carefully. The agent of the present invention supports the construction and maintenance of such lists, and facilitates comparison of candidates in the lists by various criteria.

Offer Negotiation: The shopping agent is not limited to providing information to shoppers. Negotiate prices and service options with retailers.

Product Selection, Purchasing, and Product Support: The present invention can be used as a channel to facilitate the transaction itself and provide product services.

FIG. 27B is a diagram of a mobile portal platform 2710 including a mobile portal 2412 and an electronic bullet 2713. The electronic bullet 2713
It includes a supporting hardware device 2716, such as a wireless PDA, and a mobile portal thin client standard 2714 running on a thin client operating system 2718. The mobile portal is a mobile portal server 272 that includes an encryption and decryption element 2720 and an intelligent agent 2724.
2, a customer intelligent element 2726, and a customer database 2728.

A thin client refers to reduced hardware costs, reduced maintenance and support costs, reduced LAN / WAN bandwidth requirements, reduced downtime, improved performance, and enhancements. A general term used to describe a rapidly emerging technology group that, in combination with security provided, offers a reduction in overall cost of ownership. "Thin (
The term "thin" refers to the (very small) size of the client operating system. In contrast, traditional PC operating systems (DOS,
Windows 95, etc.) are considered "fat" clients due to their large size and resource requirements. Despite the fact that the thin client operating system is thin, the thin client's capabilities are robust. Client solutions are currently deployed in mission-critical environments, providing reliable and responsive access to countless applications. Mobile portal thin client 2714 is a thin client where most of the processing is performed by mobile portal server 2722 and associated third party content and service providers 2730. Using the mobile portal thin client application 2714, the user selects a service provided by the mobile portal platform 2710 and reviews the information. The mobile portal thin client application 2714 uses a thin client operating system 2718 to make many devices independent. The thin client operating system 2718 includes a mobile portal thin client application 2714 and supporting hardware 2714.
Works as a messenger between 16 The thin client operating system 2718 allows the mobile portal thin client 2714 to make function calls to the thin client operating system 2718 for low-level hardware operations such as display calls and user input queries. A separate thin client operating system 2718 can be developed for each hardware device 2716 used as supporting hardware for the electronic bullet 2713. This allows the mobile portal thin client application 2714 to run on different supporting hardware 2716 without requiring low-level design modifications.

The mobile portal 2712 receives data from the electronic bullet 2727 via the packet-switched wireless network 2732. Information received through the packet-switched wireless network is then decoded by the encryption and decryption element 2720 of the mobile portal 2712. Once the data is decoded, mobile portal server 2722 utilizes intelligent agent 2724, customer intelligence 2726, and customer data 2728 to obtain the requested data from third party content and service provider 2730. The mobile portal server 2722 utilizes intelligent software agents to respond to customer needs. Software Agent 2722
Utilizing customer data 2728, decisions are made to personalize tasks according to the goals, habits, and preferences of individual users. Next, customer data 2728 is
Updated as usual by the customer and customer behavior. Each time a user uses the mobile portal 2712, a log is logged regarding the user's queries and other uses of the mobile portal platform 2710. This allows software agent 2724 to personalize the task using the user's past habits.

In addition to software agent 2724, mobile portal server 2722
Uses customer intelligence 2726 to respond to the needs of the user. The user can use data mining and pattern recognition to find desired information. Again, the customer data 2728 is updated to reflect the user's mining of data and the use of pattern recognition. Third party content and service providers 2730 are used by mobile portal 2712 to provide services and information requested by users. Third party content and service providers can be accessed through the Internet or mobile portal extranet. Intelligent agent software 2712 searches for third party providers,
The most suitable one for the user is determined in consideration of the customer profile included in the customer data 2728. This allows the user to reduce specificity in the query compared to what would have to be done without a user profile. For example, a user may request a jacket using mobile portal platform 2710. The intelligent agent then
The customer data 2728 is used to more specifically determine what the customer actually wants. In this case, customer data 2728 indicates that this particular user prefers denim jacket over leather jacket. Intelligent agent 2724 will then search for the denim jacket. Of course, the user can override the user profile to obtain the opposite information as stored in the user profile. Some of the typical services offered include geolocation, audio and image editing, personal news and entertainment, personal shopping, personal health and safety, personal organizers,
Includes personal finance and personal communications.

A geographic location service is typically based on information received from an integrated global positioning system. GPS data is combined with specific user request data to provide location-specific information to the user. For example, suppose a user is in San Francisco and wants to get information on good food in this city. This user uses the electronic bullet 2713 to request information on high-quality meals. Integrated G
The location data obtained from the PS receiver is combined with the user's request for a good meal, and then the combined message is sent to the mobile portal 271.
2 is transmitted. Based on the data received, the mobile portal will select the appropriate service and in this case transmit a request for a good meal in San Francisco. The mobile portal then sends the received response back to the electronic bullet 2713. The user is then presented with the requested information and formatted and displayed on the electronic bullet 2713 display device.

[0297] Audio and image editing services are typically based on data received from an integrated camera and microphone. Users typically utilize an integrated digital camera to capture images. However, the user has access to the scanner, email,
Digital images can also be obtained from other sources, such as page b. In addition, users typically utilize an integrated microphone to capture audio files. However, audio files can also be obtained from other sources, such as e-mail, web pages, and CDs. Image and / or audio data is combined with specific user requirements to provide the user with image and audio editing capabilities. For example, a user captures an image with an integrated digital camera and then
You can request that the image be edited using a particular photo editor. The image captured by the integrated digital camera is then combined with the user's request for photo editing, and the combined message is then transmitted to the mobile portal 2712. Based on the data received, the mobile portal 2712
Selects an appropriate service, and in this case, transmits a request for image editing. The mobile portal then sends the received response back to the electronic bullet 2713. The user is then presented with the requested information and formatted and displayed on the electronic bullet 2713 display device. In this case, the user receives an interface for image editing. The user then edits the image using the image editing user interface. The image change is treated as a request, and the mobile portal 2712 passes it to an image editing application running on a local or independent server.

Support Code According to Preferred Embodiment: The following code is the Microsoft Active according to the preferred embodiment.
Written and executed in the Server Pages environment. It consists primarily of Microsoft JScript with several database calls embedded in the code to query and store information in the database.

[0299] Intent-centric interface

Create Intention ASP Page (“intention_create.asp”)

[0301]

[Table 8]

Search the intention list ASP page (“intention_list.asp”)

[0303]

[Table 9]

Display the user intention list ASP page (excerpt from “navigation.asp”)

[0305]

[Table 10]

While various embodiments have been described above, it should be understood that these embodiments have been described by way of example, and not by way of limitation. Accordingly, the breadth and scope of the preferred embodiment should not be limited by any of the above-described embodiments, but should be defined only in accordance with the following claims and their equivalents.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an exemplary hardware environment according to a preferred embodiment.

FIG. 1A illustrates a prior art mobile computing solution based on a web portal network.

FIG. 2 is a flowchart of a system according to a preferred embodiment;

FIG. 3 is a flowchart of a parsing unit of the system according to a preferred embodiment;

FIG. 4 is a flowchart relating to pattern matching according to a preferred embodiment.

FIG. 5 is a flowchart for a search unit according to a preferred embodiment;

FIG. 6 is a flowchart illustrating overall system processing according to a preferred embodiment;

FIG. 7 is a flowchart of topic processing according to a preferred embodiment.

FIG. 8 is a flowchart of a meeting record process according to a preferred embodiment.

FIG. 9 is a block diagram illustrating a process flow of a pocket bargain finder according to a preferred embodiment.

FIG. 10A is a block diagram and a flowchart showing logic related to creating a customized content Web page according to a preferred embodiment.

FIG. 10B is a block diagram and a flowchart showing logic related to creating a customized content Web page according to a preferred embodiment.

FIG. 11 is a flowchart illustrating detailed logic associated with searching for user-centered content according to a preferred embodiment.

FIG. 12 illustrates a data model of a user profile according to a preferred embodiment.

FIG. 13 illustrates a persona data model according to a preferred embodiment.

FIG. 14 illustrates an intention data model according to a preferred embodiment.

FIG. 15 is a flowchart of a process relating to creation of current statistics of an agent according to a preferred embodiment;

FIG. 16 is a flow chart of logic for determining a personalized product rating for a user according to a preferred embodiment.

FIG. 17 is a flowchart of logic for accessing a centrally stored profile according to a preferred embodiment.

FIG. 18 is a flowchart of the interaction logic between a user and a specific supplier integrator according to a preferred embodiment.

FIG. 19 is a flowchart of an agent process for generating an oral summary according to a preferred embodiment.

FIG. 20 illustrates a display login according to a preferred embodiment.

FIG. 21 illustrates a daily logistics management display according to a preferred embodiment.

FIG. 22 illustrates a user main display according to a preferred embodiment.

FIG. 23 illustrates an agent interaction display according to a preferred embodiment.

FIG. 24 is a block diagram illustrating an active knowledge management system according to a preferred embodiment.

FIG. 25 is a block diagram illustrating a back-end server according to a preferred embodiment.

FIG. 26 is a flowchart illustrating how the hardware and software of one embodiment of the present invention operate.

FIG. 27A illustrates a display in a browser mode according to a preferred embodiment.

FIG. 27B illustrates a mobile portal platform according to a preferred embodiment.

[Explanation of symbols]

 110 ... CPU 116 ... ROM 114 ... RAM 112 ... Bus 118 ... I / O adapter 120 ... Disk storage unit 124 ... Keyboard 126 ... Mouse 128 ... Speaker 132 ... Microphone 134 ... Communication adapter 135 ... Network 136 ... Display adapter 138 ... Display device 1010 … Customer web browser 1020… Integrator web server 1030… Intent database 1040… Content database 1050… Supplier profile database 1060… Customer profile database 1070… Supplier web server 1080… Product information database 1090… Product order database 1091… Customer information database 1710… Profileデ ー タ ベ ー ス Database 1720 Profile gateway server 1730 パ ブ リ ッ ク Public Internet 1740 テ レ ビ Television 1750 Trader 2000, 2220, 2310 エ ー ジ ェ ン ト Agent 2030 ロ ゴ Logo 2040 ボ タ ン Button 2100 ア ニ メ ー シ ョ ン Animation agent 2110 旅行 Travel 2120 家庭 Household chores 2130 金融 Finance 2140 市場 Market activity 2142 ... icon 2146 ... direct link 2150 ... permission 2180 ... item 2200 ... world news 2210 ... local news 2230 ... real estate 2300 ... information 2370 ... chart 2380 ... button 2390 ... set of buttons 2395 ... hint 2400 ... backend 2410, 2420 ... magic Wall 2430, 2436… Awareness machine 2446, 2448 Web client 2450, 2460 E-mail client 2710 Mobile portal platform 2712 Mobile portal 2713 Electronic bullet 2714 Thin client 2716 Hardware device 2718 Operating system 2720 Encryption and decryption element 2722 Mobile Portal Server 2724 Intelligent Agent 2726 Customer Intelligent Element 2728 Customer Database 2730 Third Party Content and Service Provider 2732 Packet Switched Wireless Network

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme court ゛ (Reference) G06F 15/02 340 G06F 15/02 340Z (81) Designated country EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE), OA (BF, BJ, CF, CG, CI, CM, GA, GN, GW, ML, MR, NE, SN, TD, TG), AP (GH, GM, KE, LS, MW, SD, SL, SZ, TZ, UG, ZW), EA (AM, AZ, BY, KG, KZ, MD) , RU, TJ, TM), AE, AL, AM, AT, AU, AZ, BA, BB, BG, BR, BY, CA, CH, CN, CR, CU, CZ DE, DK, DM, EE, ES, FI, GB, GD, GE, GH, GM, HR, HU, ID, IL, IN, IS, JP, KE, KG, KP, KR, KZ, LC, LK , LR, LS, LT, LU, LV, MA, MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE, SG, SI, SK, SL, TJ, TM, TR, TT, TZ, UA, UG, UZ, VN, YU, ZA, ZW (72) Inventor Swaminathan Kishore Sandaram 60056 Downers Grove, Stair Street, Illinois United States 6524 (72) Inventor Myers James L. 6014 Chicago, Illinois, USA, Apartment # 1, West Lightwood, 1107 (72) Inventor Fano Andrew E. 60202 Evanston, Illinois, United States Apartment 1E, Maple Avenue, 1137 F-term (reference) 5B019 GA03 GA10 HG30

Claims (19)

[Claims] 1. A method for obtaining health care on a mobile computing environment, comprising: (a) creating a query based in part on user input; and (b) an information network utilizing a wireless communication device. (C) receiving a response to the inquiry from the information network on the mobile computing environment; (d) monitoring health information appropriate to a user; (e) an application tool Processing the health information and response utilizing: (f) displaying health information to a user. 2. The method of obtaining health care on a mobile computing environment according to claim 1, wherein the health information includes nutrition information. 3. The method of obtaining health care on a mobile computing environment according to claim 1, comprising monitoring a user's exercise program and displaying recommended exercises. 4. The method of obtaining health care on a mobile computing environment according to claim 1, comprising displaying self-diagnosis information. 5. The method of obtaining health care on a mobile computing environment according to claim 1, comprising displaying an emergency message and transmitting the emergency message. 6. The method of obtaining health care on a mobile computing environment according to claim 1, comprising coordinating health records with medical personnel. 7. The method of obtaining health care on a mobile computing environment according to claim 1, comprising adjusting insurance claims. 8. The method of obtaining health care on a mobile computing environment according to claim 1, comprising displaying information of special conditions. 9. The method of obtaining health care on a mobile computing environment according to claim 1, comprising coordinating post-treatment care for a user by a medical facility. 10. An apparatus for providing health care information on a mobile computing environment, comprising: (a) a processor; (b) a memory for storing information under control of the processor; Logic to create a query based in part; (d) logic to query an information network using a wireless communication device; (e) logic to monitor health information appropriate to the user; and (f) application tools. An apparatus comprising: logic for processing the health information and the response using the logic; and (g) logic for displaying health information to a user. 11. A computer program embedded on a computer readable medium for providing healthcare information on a mobile computing environment, comprising: (a) a code segment for creating a query based in part on user input; (B) a code segment for querying an information network using a wireless communication device; (c) a code segment for receiving a response to the query from the information network on the mobile computing environment; (E) a code segment that processes the health information and the response using an application tool; and (f) a code segment that displays health information to a user. Computer program. 12. The computer program of claim 11, wherein the health information comprises nutrition information, embedded on a computer readable medium and providing health care on a mobile computing environment. 13. The computer program of claim 11, wherein the computer program is embedded on a computer-readable medium to provide health care on a mobile computing environment and monitors a user's exercise program to indicate recommended exercise. Computer program containing code segments to be executed. 14. The computer program of claim 11, wherein the computer program is embedded on a computer readable medium and provides health care on a mobile computing environment, comprising a code segment displaying self-diagnosis information. 15. The computer program of claim 11, which is embedded on a computer readable medium and provides health care on a mobile computing environment, the computer program for displaying an emergency message and transmitting the emergency message. Computer programs, including. 16. The computer program of claim 11 embodied on a computer readable medium to provide health care on a mobile computing environment, comprising a code segment that coordinates health records by medical personnel. Computer program. 17. The computer program of claim 11, wherein the computer program is embedded on a computer readable medium and provides health care on a mobile computing environment, comprising a code segment that coordinates insurance claims. 18. The computer program of claim 11, wherein the computer program is embedded on a computer readable medium and provides health care on a mobile computing environment, comprising a code segment displaying information of a special condition. program. 19. The computer program of claim 11, wherein the computer program is embedded on a computer readable medium and provides health care on a mobile computing environment, coordinates post-treatment care for a user by a healthcare facility. A computer program containing code segments.