Multi-Lingual Knowledge Matrix Method and System

by Go Simon Sunatori, P.Eng./ing., M.Eng. (Engineering Physics), F.N.A.

A Thesis (Discourse) Submitted to the Institute of Information Technology in Partial Fulfilment of the Requirements for the Degree Renaissance Doctorate (R.D.)

New York College of Advanced Studies (NYCAS)

2003-06

Abstract
A multi-lingual knowledge matrix having knowledge packets represents mutually-exclusive knowledge at high level of abstraction. One aspect of the innovation is a method for encapsulating multi-lingual knowledge packets of an element and its components in a multi-lingual knowledge matrix having a plurality of dimensions. Another aspect of the innovation is a multi-lingual system having language-independent algorithms to compile knowledge packets by a correct-by-construction matrix operation. The knowledge matrix compiler comprises a language processor which separates languages, and a knowledge packet parser which processes knowledge by concatenating lines, sorting lines, removing duplicates, and parsing selectively.

Abstract

A multi-lingual knowledge matrix having knowledge packets represents mutually-exclusive knowledge at high level of abstraction. One aspect of the innovation is a method for encapsulating multi-lingual knowledge packets of an element and its components in a multi-lingual knowledge matrix having a plurality of dimensions. Another aspect of the innovation is a multi-lingual system having language-independent algorithms to compile knowledge packets by a correct-by-construction matrix operation. The knowledge matrix compiler comprises a language processor which separates languages, and a knowledge packet parser which processes knowledge by concatenating lines, sorting lines, removing duplicates, and parsing selectively.

Introduction

Description

Title
Technical Field
Background Art
Description of the Invention
Brief Description of the Figures in the Drawings
Modes for Carrying Out the Invention

Bibliography/References

Introduction

The original Knowledge Matrix Method and System was conceived in order to minimise overlap, duplication and redundancy in information technology systems in the spirit of absolute simplicity and ultimate consistency. This invention provides means for effectively organising source knowledge in order to automatically generate object information in a push-button operation.
The Multi-Lingual Knowledge Matrix Method and System represents a paradigm shift from language translation to coding multi-lingual knowledge packets in correct-by-construction method. The technique has potential for useful applications in describing structured contents in the form of Subject-Verb-Complement.
One example of such real-life applications is the Food Nutrition Knowledge Matrix. The database stores knowledge packets coded in "English|French" format, and the software generates English and French versions of readable text by knowledge matrix operations.
The Food Nutrition Knowledge Matrix is just one example of potential applications of this invention. The Multi-Lingual Knowledge Matrix Method and System may lead to many possible commercial products, such as Herb Nutrition Knowledge Matrix and Pharmaceutical Drug Knowledge Matrix, as well as to internal process improvements in corporations and governments, for example.
An Apple Macintosh computer and HyperCard software were used in conceiving ideas, developing prototypes, applying, and documenting the invention.
Although the software application is written in the HyperTalk language, the method and system are generic enough to apply to any topic (other than Food and Nutrition) on any computing platform (other than Apple Macintosh) in any programming language (other than HyperTalk).

Description

Title

Multi-Lingual Knowledge Matrix Method and System

Technical Field

This invention relates to computing, more particularly to digital data processing systems and methods.

Background Art

There are many instances where it would be desirable to be able to provide a method and system to compile knowledge in a plurality of languages from a single source.
A Canadian patent discloses a uni-lingual knowledge matrix method and system. There are other knowledge base management methods and systems using multiple knowledge bases.
These prior art arrangements do not accommodate more than one language at the same time in a single knowledge base. The systems that deal with more than one language employ methods to translate and to maintain multiple knowledge bases, i.e., one knowledge base per each language.

Description of the Invention

It is a primary object of the invention to provide a method to code multi-lingual knowledge in a modular fashion.
It is another object of the invention to provide a system to compile uni-lingual knowledge and produce output from a multi-lingual knowledge base based on an input query.
One aspect of the invention provides a concept of multi-lingual knowledge packet to represent mutually-exclusive knowledge at higher level of abstraction than a pixel, a letter, and a word, in ascending order. Each multi-lingual knowledge packet is coded in table format in a knowledge base. The multi-lingual knowledge packet for an element comprises a list of components, while the multi-lingual knowledge packet for a component has a Subject-Verb-Complement form to encapsulate knowledge.

The following example shows multi-lingual knowledge packets in 2 languages, English and French, being separated by a vertical bar "|" as a delimiter.

------------------------------------------------------------------------
[Element|Élément]

Components|Composants
------------------------------------------------------------------------
Component (Subject)|Composant (Sujet)

Properties|Propriétés
* Verb:|Verbe:
  * Complement|Complément
------------------------------------------------------------------------

Another aspect of the invention provides a method for encapsulating multi-lingual knowledge packets of an element in a multi-lingual knowledge matrix having a plurality of dimensions, i.e., (N X N) and (N X 1) where N is a positive integer. The following example shows multi-lingual knowledge matrices having dimensions of (2 X 2) and (2 X 1).
```
------------------------------------------------------------------------
[Element|Élément]

= [Property.1|Propriété.1 Property.1|Propriété.1] * [Component.1|Composant.1]
  [Property.2|Propriété.2 Property.2|Propriété.2]   [Component.2|Composant.2]
------------------------------------------------------------------------
```

Another aspect of the invention provides a system having language-independent algorithms to compile multi-lingual knowledge packets of components by means of a correct-by-construction matrix transformation. Thus, the multi-lingual knowledge matrices in this example are now transformed into a single multi-lingual knowledge matrix having a dimension of (2 X 1). The single multi-lingual knowledge matrix can now be separated into one uni-lingual matrix per each language.

------------------------------------------------------------------------
[Element]

= [(Property.1 * Component.1) + (Property.1 * Component.2)]
  [(Property.2 * Component.1) + (Property.2 * Component.2)]
------------------------------------------------------------------------
[Élément]

= [(Propriété.1 * Composant.1) + (Propriété.1 * Composant.2)]
  [(Propriété.2 * Composant.1) + (Propriété.2 * Composant.2)]
------------------------------------------------------------------------

Brief Description of the Figures in the Drawings

In drawings which illustrate embodiments of the invention:
- Figure 1 is a block diagram of a multi-lingual knowledge matrix system;
- Figure 2 is a flow chart of a multi-lingual knowledge matrix compiler; and
- Figure 3 is a flow chart of a multi-lingual knowledge packet parser.

Modes for Carrying Out the Invention

Embodiments of the invention will now be described by way of example using a food element "[Garlic|Ail]" in the "Food Nutrition Analyser". The properties of the foods are based solely on their component nutrients.
In this example, the food element "[Garlic|Ail]" has 3 components: "Ajoene|Ajoene", "Allicin|Allicin", and "SAC|SAC". There are 3 properties of food nutrition: "General|Générales", "Helpful|Utiles", and "Harmful|Nocives".
According to the present invention shown in the block diagram of Figure 1, the multi-lingual knowledge matrix system comprises a Multi-Lingual Knowledge Base 1-1 and a Multi-Lingual Knowledge Matrix Compiler 1-2. The Multi-Lingual Knowledge Matrix Compiler 1-2 has a Language Processor 1-0 and a Knowledge Packet Parser 1-3. The multi-lingual knowledge matrix system takes an Input Query, and compiles knowledge from the Multi-Lingual Knowledge Base 1-1 to produce Output Knowledge. In one embodiment of the invention shown in Figure 1, the Output Knowledge is in English and in French.

In the Multi-Lingual Knowledge Base 1-1, the multi-lingual knowledge packet to represent the food element "[Garlic|Ail]" is coded in table format as Subject-Verb-Complement, as follows. In this example, the multi-lingual knowledge packets have 2 languages, English and French, being separated by a vertical bar "|" as a delimiter.

------------------------------------------------------------------------
[Garlic|Ail]

Components|Composants
* Ajoene|Ajoene
* Allicin|Allicin
* SAC|SAC

General|Générales
* Is:|Est:
  * an allium family vegetable|un légume de la famille de l'allium
------------------------------------------------------------------------

The multi-lingual knowledge packet to represent each component of the food element "[Garlic|Ail]" is coded in table format as Subject-Verb-Complement, as follows.

------------------------------------------------------------------------ Ajoene|Ajoene

The food element "[Garlic|Ail]" is now represented by 2 multi-lingual knowledge matrices having dimensions of (3 X 3) and (3 X 1), as follows.

------------------------------------------------------------------------
[Garlic|Ail]

= [General|Générales General|Générales General|Générales] * [Ajoene|Ajoene  ]
  [Helpful|Utiles    Helpful|Utiles    Helpful|Utiles   ]   [Allicin|Allicin]
  [Harmful|Nocives   Harmful|Nocives   Harmful|Nocives  ]   [SAC|SAC        ]
------------------------------------------------------------------------

The multi-lingual knowledge matrices are filtered into separate uni-lingual matrices by a language processor, in English and in French. For each language, the uni-lingual knowledge matrices can now be transformed to produce a single uni-lingual knowledge matrix for each language having a dimension of (3 X 1), as shown below.

------------------------------------------------------------------------
[Garlic]

= [(General * Ajoene) + (General * Allicin) + (General * SAC)]
  [(Helpful * Ajoene) + (Helpful * Allicin) + (Helpful * SAC)]
  [(Harmful * Ajoene) + (Harmful * Allicin) + (Harmful * SAC)]
------------------------------------------------------------------------
[Ail]

= [(Générales * Ajoene) + (Générales * Allicin) + (Générales * SAC)]
  [(Utiles    * Ajoene) + (Utiles    * Allicin) + (Utiles    * SAC)]
  [(Nocives   * Ajoene) + (Nocives   * Allicin) + (Nocives   * SAC)]
------------------------------------------------------------------------

The language-independent knowledge matrix transformation concatenates lines, sorts lines, removes duplicates, and parses selectively to transform the source knowledge into object information in English and in French, as shown below.

------------------------------------------------------------------------ [Garlic]

This Element contains the following Components: * Ajoene * Allicin * SAC

General Properties of the Components: * Is: * an allium family vegetable

Helpful Properties of the Components: * Inhibits clotting of blood. * Lowers: * level of blood cholesterol * level of blood pressure * Prevents: * heart attack * infection * stroke ------------------------------------------------------------------------ [Ail]

Cet Élément contient les Composants suivants: * Ajoene * Allicin * SAC

Propriétés Générales des Composants: * Est: * un légume de la famille de l'allium

Propriétés Utiles des Composants: * Empêche la coagulation du sang. * Prévient: * attaque * crise cardiaque * infection * S'abaisse: * niveau de cholestérol dans le sang * niveau de tension artérielle ------------------------------------------------------------------------

Figure 2 shows a flow chart of the Multi-Lingual Knowledge Matrix Compiler 1-1 in Figure 1. The flow starts with a recursive process to Collect Components 2-1 used in an element, and a recursive process to Collect Properties 2-2, followed by a process to determine whether an item is an Element or a Component? 2-3. If the item is a component, the system goes through an extra process to Collect Elements 2-4 used by a component. After the multi-lingual knowledge packets are all assembled, the Filter Language 2-0 stage turns the multi-lingual packets into uni-lingual knowledge packets, and the Parse Knowledge Packet 2-5 stage processes them for the final step to Present Output 2-6.

Each section of the HyperTalk code shown below implements the algorithm of the process.

------------------------------------------------------------------------
function knowledgeMatrix n,theLanguage
  set the cursor to "busy"
  put "" into theTitle
  put "" into theGeneral
  put "" into theHelpful
  put "" into theHarmful
  put "" into theScript
  if (there is a cd n) then
    put the itemDelimiter into oldDelimiter
    --
    -- Collect Components used in an Element, Collect Properties
    --
    if (bg fld "General" of cd n <> "") then
      put (bg fld "General" of cd n) & return after theGeneral
    end if
    if (bg fld "Helpful" of cd n <> "") then
      put (bg fld "Helpful" of cd n) & return after theHelpful
    end if
    if (bg fld "Harmful" of cd n <> "") then
      put (bg fld "Harmful" of cd n) & return after theHarmful
    end if
    repeat with i = 1 to the number of lines of bg fld "Contents" of cd n
      put line i of bg fld "Contents" of cd n into o
      if (there is a cd o) then
        if ((first char of the short name of cd o <> "{") or (last char of the short name of cd o <> "}")) then
          put (bg fld "Title" of cd o) & return after theTitle
        end if
        if (bg fld "General" of cd o <> "") then
          put (bg fld "General" of cd o) & return after theGeneral
        end if
        if (bg fld "Helpful" of cd o <> "") then
          put (bg fld "Helpful" of cd o) & return after theHelpful
        end if
        if (bg fld "Harmful" of cd o <> "") then
          put (bg fld "Harmful" of cd o) & return after theHarmful
        end if
        repeat with j = 1 to the number of lines of bg fld "Contents" of cd o
          put line j of bg fld "Contents" of cd o into p
          if (there is a cd p) then
            if ((first char of the short name of cd p <> "{") or (last char of the short name of cd p <> "}")) then
              put (bg fld "Title" of cd p) & return after theTitle
            end if
            if (bg fld "General" of cd p <> "") then
              put (bg fld "General" of cd p) & return after theGeneral
            end if
            if (bg fld "Helpful" of cd p <> "") then
              put (bg fld "Helpful" of cd p) & return after theHelpful
            end if
            if (bg fld "Harmful" of cd p <> "") then
              put (bg fld "Harmful" of cd p) & return after theHarmful
            end if
            repeat with k = 1 to the number of lines of bg fld "Contents" of cd p
              put line k of bg fld "Contents" of cd p into q
              if (there is a cd q) then
                if ((first char of the short name of cd q <> "{") or (last char of the short name of cd q <> "}")) then
                  put (bg fld "Title" of cd q) & return after theTitle
                end if
                if (bg fld "General" of cd q <> "") then
                  put (bg fld "General" of cd q) & return after theGeneral
                end if
                if (bg fld "Helpful" of cd q <> "") then
                  put (bg fld "Helpful" of cd q) & return after theHelpful
                end if
                if (bg fld "Harmful" of cd q <> "") then
                  put (bg fld "Harmful" of cd q) & return after theHarmful
                end if
              end if
            end repeat
          end if
        end repeat
      end if
    end repeat
    --
    -- Collect Elements used by a Component
    --
    if ((first char of the short name of cd n <> "[") or (last char of the short name of cd n <> "]")) then
      repeat with i = 3 to the number of cds
        if (return & (bg fld "Title" of cd n) & return is in return & (bg fld "Contents" of cd i) & return) then
          if ((first char of the short name of cd i = "{") and (last char of the short name of cd i = "}")) then
            repeat with j = 3 to the number of cds
              if (return & (bg fld "Title" of cd i) & return is in return & (bg fld "Contents" of cd j) & return) then
                put (bg fld "Title" of cd j) & return after theScript
              end if
            end repeat
          else
            put (bg fld "Title" of cd i) & return after theScript
          end if
        end if
      end repeat
    end if
    --
    -- Filter Language, Parse Knowledge Packet and Present Output
    --
    set the itemDelimiter to "|"
    put return into bg fld "Script" of cd n
    if (theTitle <> "") then
      put \
      "This Elements contains the following Components:" & "|" & \
      "Cet Élément contient les Composants suivants:" into theHeading
      put (item theLanguage of theHeading) & return & return & \
      knowledgePacket(theTitle,theLanguage) & return & \
      "------------------------------------------------------------------------" & return & return after bg fld "Script" of cd n
    end if
    if (theGeneral <> "") then
      put \
      "General Properties of the Components:" & "|" & \
      "Propriétés Générales des Composants:" into theHeading
      put (item theLanguage of theHeading) & return & return & \
      knowledgePacket(theGeneral,theLanguage) & return & \
      "------------------------------------------------------------------------" & return & return after bg fld "Script" of cd n
    end if
    if (theHelpful <> "") then
      put \
      "Helpful Properties of the Components:" & "|" & \
      "Propriétés Utiles des Composants:" into theHeading
      put (item theLanguage of theHeading) & return & return & \
      knowledgePacket(theHelpful,theLanguage) & return & \
      "------------------------------------------------------------------------" & return & return after bg fld "Script" of cd n
    end if
    if (theHarmful <> "") then
      put \
      "Harmful Properties of the Components:" & "|" & \
      "Propriétés Nocif des Composants:" into theHeading
      put (item theLanguage of theHeading) & return & return & \
      knowledgePacket(theHarmful,theLanguage) & return & \
      "------------------------------------------------------------------------" & return & return after bg fld "Script" of cd n
    end if
    if (theScript <> "") then
      put \
      "This Component is contained in the following Elements:" & "|" & \
      "Cet Composant est contenu dans les Éléments suivants:" into theHeading
      put (item theLanguage of theHeading) & return & return & \
      knowledgePacket(theScript,theLanguage) & return & \
      "------------------------------------------------------------------------" & return & return after bg fld "Script" of cd n
    end if
    repeat with i = 1 to the number of lines of bg fld "Script" of cd n
      if      (char 1 of line i of bg fld "Script" of cd n = "-") then
        set the textSize  of line i of bg fld "Script" of cd n to "10"
        set the textStyle of line i of bg fld "Script" of cd n to "plain"
      else if (char 1 of line i of bg fld "Script" of cd n = "�") then
        set the textSize  of line i of bg fld "Script" of cd n to "10"
        set the textStyle of line i of bg fld "Script" of cd n to "bold"
      else if (char 1 of line i of bg fld "Script" of cd n <> " ") then
        set the textSize  of line i of bg fld "Script" of cd n to "12"
        set the textStyle of line i of bg fld "Script" of cd n to "bold"
      else
        set the textSize  of line i of bg fld "Script" of cd n to "10"
        set the textStyle of line i of bg fld "Script" of cd n to "plain"
      end if
    end repeat
    --
    set the itemDelimiter to oldDelimiter
  end if
end knowledgeMatrix
------------------------------------------------------------------------

Figure 3 shows a flow chart of the Language Processor 1-0 and the Knowledge Packet Parser 1-2 in Figure 1. In this example, the input is the Helpful Properties of all components of the food element "[Garlic|Ail]", as shown below.

------------------------------------------------------------------------
* Inhibits:|Empêche:
  * clotting of blood|la coagulation du sang
* Prevents:|Prévient:
  * heart attack|attaque 
  * stroke|crise cardiaque
* Inhibits:|Empêche:
  * clotting of blood|la coagulation du sang
* Lowers:|S'abaisse:
  * level of blood cholesterol|niveau de cholestérol dans le sang
  * level of blood pressure|niveau de tension artérielle
* Prevents:|Prévient:
  * infection|infection
------------------------------------------------------------------------

The Separate Languages 3-0 process separates the multi-lingual text into uni-lingual English text and uni-lingual French text, as shown below.

------------------------------------------------------------------------
* Inhibits:
  * clotting of blood
* Prevents:
  * heart attack
  * stroke
* Inhibits:
  * clotting of blood
* Lowers:
  * level of blood cholesterol
  * level of blood pressure
* Prevents:
  * infection
------------------------------------------------------------------------
* Empêche:
  * la coagulation du sang
* Prévient:
  * attaque 
  * crise cardiaque
* Empêche:
  * la coagulation du sang
* S'abaisse:
  * niveau de cholestérol dans le sang
  * niveau de tension artérielle
* Prévient:
  * infection
------------------------------------------------------------------------

The Concatenate Lines 3-1 process concatenates the verb of each property with its complements to form a list.

------------------------------------------------------------------------
* Inhibits:clotting of blood
* Prevents:heart attack
* Prevents:stroke
* Inhibits:clotting of blood
* Lowers:level of blood cholesterol
* Lowers:level of blood pressure
* Prevents:infection
------------------------------------------------------------------------
* Empêche:la coagulation du sang
* Prévient:attaque 
* Prévient:crise cardiaque
* Empêche:la coagulation du sang
* S'abaisse:niveau de cholestérol dans le sang
* S'abaisse:niveau de tension artérielle
* Prévient:infection
------------------------------------------------------------------------

The Sort Lines 3-2 process sorts the list in ascending alphabetical order.

------------------------------------------------------------------------
* Inhibits:clotting of blood
* Inhibits:clotting of blood
* Lowers:level of blood cholesterol
* Lowers:level of blood pressure
* Prevents:heart attack
* Prevents:infection
* Prevents:stroke
------------------------------------------------------------------------
* Empêche:la coagulation du sang
* Empêche:la coagulation du sang
* Prévient:attaque 
* Prévient:crise cardiaque
* Prévient:infection
* S'abaisse:niveau de cholestérol dans le sang
* S'abaisse:niveau de tension artérielle
------------------------------------------------------------------------

The Remove Duplicates 3-3 process removes duplicate lines from the list.

------------------------------------------------------------------------
* Inhibits:clotting of blood
* Lowers:level of blood cholesterol
* Lowers:level of blood pressure
* Prevents:heart attack
* Prevents:infection
* Prevents:stroke
------------------------------------------------------------------------
* Empêche:la coagulation du sang
* Prévient:attaque 
* Prévient:crise cardiaque
* Prévient:infection
* S'abaisse:niveau de cholestérol dans le sang
* S'abaisse:niveau de tension artérielle
------------------------------------------------------------------------

The process to Parse Selectively 3-3 analyses lines in context. If there is only one line that contains a particular verb, the verb is paired with the corresponding complement, and the colon ":" is substituted with a space " " in order to build a sentence which is terminated with a period ".". If there are more than one line that contain a particular verb, the verb is terminated with a colon ":", and a list of complements follow in order to reconstruct a sentence.

------------------------------------------------------------------------
* Inhibits clotting of blood.
* Lowers:
  * level of blood cholesterol
  * level of blood pressure
* Prevents:
  * heart attack
  * infection
  * stroke
------------------------------------------------------------------------
* Empêche:
  * la coagulation du sang
* Prévient:
  * attaque 
  * crise cardiaque
  * infection
* S'abaisse:
  * niveau de cholestérol dans le sang
  * niveau de tension artérielle
------------------------------------------------------------------------

Each section of the HyperTalk code shown below implements the algorithm of the process.

------------------------------------------------------------------------
function knowledgePacket theMulti,theLanguage
  put "" into theText
  if (theMulti <> "") then
    put the itemDelimiter into oldDelimiter
    --
    -- Separate Languages
    --
    set the itemDelimiter to "|"
    put "" into theUni
    repeat with i = 1 to the number of lines of theMulti
      put line i of theMulti into theLine
      if      (char 1 to 2 of theLine = "� ") then
        put (char 3 to (the number of chars of theLine) of theLine) into theLine
        put "� "   & (item theLanguage of theLine) & return after theUni
      else if (char 1 to 4 of theLine = "  � ") then
        put (char 5 to (the number of chars of theLine) of theLine) into theLine
        put "  � " & (item theLanguage of theLine) & return after theUni
      else if (("[" is in theLine) and ("]" is in theLine)) then
        delete first char of theLine
        delete last  char of theLine
        put "� [" & (item theLanguage of theLine) & "]" & return after theUni
      else if (theLine <> "") then
        put "� "  & (item theLanguage of theLine)       & return after theUni
      end if
    end repeat
    --
    -- Concatenate Lines with ":"
    --
    set the itemDelimiter to ":"
    put "" into theSort
    repeat with i = 1 to the number of lines of theUni
      put line i of theUni into theI
      if (theI <> "") then
        if (":" is in theI) then
          put item 1 of theI into the1
        else
          if (char 1 of theI = " ") then
            put the1 & ":" & theI into theI
          end if
          put theI & return after theSort
        end if
      end if
    end repeat
    --
    -- Sort Lines, Remove Duplicates, Parse Selectively with ":" or "."
    --
    sort theSort ascending international
    repeat with i = 1 to the number of lines of theSort
      put line i       of theSort into theI
      put line (i + 1) of theSort into theJ
      if (theI <> theJ) then
        if (":" is in theI) then
          put item 1 of theI into the1
          put item 1 of theJ into the2
          if      ((the1 <> the0) and (the1 <> the2)) then
            put (item 1 of theI) & " " & (char 5 to (the number of chars of item 2 of theI) of (item 2 of theI)) & "." & return after theText
          else if ((the1 <> the0) and (the1 = the2)) then
            put the1 into the0
            put (item 1 of theI) & ":" & return after theText
            put (item 2 of theI) & return after theText
          else
            put (item 2 of theI) & return after theText
          end if
        else
          put theI & return after theText
        end if
      end if
    end repeat
    --
    set the itemDelimiter to oldDelimiter
  end if
  return theText
end knowledgePacket
------------------------------------------------------------------------

Example

Description

HyperBible of Healthy and Unhealthy Foods!
There are thousands of electronic resources, books and research papers on foods and nutrients. Their effectiveness is limited, however, because of repetitive linear presentation with redundancy and duplication, causing information overload in your brain.
The Food Nutrition Knowledge Matrix organises hundreds of foods and nutrients in a simple and consistent manner. It focuses on only major helpful and harmful food nutrients, such as minerals, vitamins, lipids and other important compounds, rather than on calories. The properties of the foods are based solely on their nutrients, which are accessible with the click of a button.

Software Preview

Some of the screen shots of the Food Nutrition Knowledge Matrix software are shown below.
Index Page
Input Source Knowledge | Food (Garlic)
Input Source Knowledge | Nutrient (Ajoene)
Input Source Knowledge | Nutrient (Allicin)
Output Object Information [English]
Output Object Information [French]

Obtain the Software

The Food Nutrition Knowledge Matrix is a stand-alone application for the Apple Macintosh computer, but it will be converted to a much more compact and cross-platform QuickTime Interactive (QTi) format. It will be on-line playable directly from the web browser and downloadable to any personal computer.

Bibliography/References

Knowledge Matrix Method and System
- Inventor(s) ............ Sunatori, Go Simon
- Patent Number .......... Canada 2 256 266
- Status ................. Granted
- Filing Date ............ 1998-12-22
- Issue Date ............. 2000-10-10
Multi-Lingual Knowledge Matrix Technique
- Inventor(s) ............ Sunatori, Go Simon
- Patent Number .......... Canada 2 260 373
- Status ................. Granted
- Filing Date ............ 1999-01-28
- Issue Date ............. 2000-12-26
Food Nutrition Knowledge Matrix
- Website URL ............ http://WWW.HyperInfo.CA/FoodNutrition/
La matrice de connaissance de la nutrition
- Website URL ............ http://WWW.HyperInfo.CA/FoodNutrition/