Multi-dimensional method and apparatus for automated language interpretation

• US 20030216919A1
• Filed: 05/13/2002
• Published: 11/20/2003
• Est. Priority Date: 05/13/2002
• Status: Active Grant

• Current Assignee: Knowledgenetica Corporation
• Original Assignee: Knowledgenetica Corporation
• Inventors: Roushar, Joseph C.
• Granted Patent: US 7,403,890 B2
• Time in Patent Office: Days
• US Class Current: 704/260
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

A method and apparatus for natural language interpretation are described. The invention includes a schema and apparatus for storing, in digital, analog, or other machine-readable format, a network of propositions formed of a plurality of text and/or non-text objects, and the steps of retrieving a string of input text, and locating all associated propositions in the network for each word in the input string. Embodiments of the invention also include optimization steps for locating said propositions, and specialized structures for storing them in a ready access storage area simulating human short-term memory. The schema and steps may also include structures and processes for obtaining and adjusting the weights of said propositions to determine posterior probabilities representing the intended meaning. Embodiments of the invention also include an apparatus designed to apply an automated interpretation algorithm to automated voice response systems and portable knowledge appliance devices.

Claims

1. A schema for storing, in digital, analog, or other machine-readable format, a proposition formed of a plurality of objects, comprising:

a molecular-style structure having a plurality of objects wherein said objects function as elements x, y, r, and c and wherein said objects form an explicit proposition comprising at least one of each x, y, r, and c element, wherein x is a first object, y is a second object, r represents a known explicit relationship between said first and second objects, c represents a known logical context of said relationship;

said relationship forming a logical proposition stating a fact regarding said first object; and

said proposition having a predetermined weight assigned thereto.

2. A structure according to claim 1, wherein said molecular-style structure is constructed and arranged such that every object is directly or indirectly linked to at least one other object, and such that an explicit proposition is identified by the links between said objects comprising the elements x, y, r, and c, and said weight, forming said explicit proposition.

3. A structure according to claim 1, wherein at least one x element is identical to another x element and wherein one identical x element is directly connected to every other identical x element.

4. The structure according to claim 1, wherein said objects are comprised of lexical and non-lexical items.

5. The structure according to claim 4, wherein said lexical items are from the group consisting of:

letters, symbols, words, numbers, and combinations thereof.

6. The structure according to claim 4, wherein said non-lexical items are from the group consisting of:

graphical items, audio items, video items, aromatic items, tactile items, and combinations thereof.

7. The structure according to claim 1, wherein said molecular-style structure is constructed and arranged such that each said object is capable of being utilized as an element x, y, r, or c with respect to at least one other object.

8. The structure according to claim 1, wherein said weight is constructed and arranged to be adjustable.

9. The structure according to claim 8, wherein said structure is constructed and arranged to store said adjusted weight in a long-term memory structure.

10. The structure according to claim 1, wherein said objects are from the group consisting of:

letters, symbols, words, numbers, and combinations thereof.

11. The structure according to claim 1, wherein said objects may further function as elements q, wherein q is a qualifier defining any known constraint that further describes the unique relationship between said first and second objects, and wherein each said object is capable of being utilized as an element x, y, r, c, or q with respect to at least one other object.

12. The structure according to claim 1, wherein said weight represents a relative measure of probability.

13. The structure according to claim 1, wherein said weight is assigned a numeric value.

14. The structure according to claim 1, wherein said plurality of objects is represented in human-readable format.

15. The structure according to claim 1, wherein said plurality of objects is encoded in a form that is not human readable.

16. The structure according to claim 1, wherein said plurality of objects is compressed in a form that is not human readable.

17. A schema for storing, in digital, analog, or other machine-readable format, a plurality of propositions formed of a plurality of text or non-text objects, comprising:

a storage architecture having a plurality of objects therein, each said object being linked within said storage architecture to another said object, said links defined by the relationship of said two objects in the context of said formal logic proposition; and

an object network structure having an arbitrarily large plurality of molecular groupings of objects in which each molecular grouping of objects represents a formal logic proposition and each said object molecule is formed of a plurality of objects, each with a specified role in forming said logical proposition.

18. A schema of claim 17, wherein said object network is a Bayesian network with each molecular grouping of objects representing one node in said Bayesian network.

19. A schema of claim 18, wherein each molecular grouping has a weight and wherein said weights of said molecular groupings of objects represent the a-priori probabilities of said nodes in said Bayesian network.

20. The structure according to claim 17, wherein said plurality of objects is represented in human-readable format.

21. The structure according to claim 17, wherein said plurality of objects is encoded in a form that is not human readable.

22. The structure according to claim 17, wherein said plurality of objects is compressed in a form that is not human readable.

23. A storage architecture for storing and processing proposition information in digital, analog, and other machine-readable formats, comprising;

a permanent storage area for retaining a body of proposition information therein;

a working storage area for temporary storage and analysis of portions of said information; and

a ready access storage area for retaining a portion of said information from said working storage area, said information in said working storage area being copied from at least one of said other storage areas.

24. The storage architecture of claim 23, wherein said body of information is organized in an object network, wherein all objects are linked to at least one other object.

25. The storage architecture of claim 23, wherein said structure further comprises a mechanism for determining the length of time since the most recent access of information in the ready access storage area.

26. The storage architecture of claim 25, said architecture further comprising means for identifying and removing information from said ready access storage area that has not been accessed for a predetermined time.

27. The storage architecture of claim 23, wherein said architecture further comprises means for identifying and manipulating information in said ready access storage area.

28. The storage architecture of claim 23, wherein said permanent storage area maintains information regarding objects, and information regarding propositions formed from a plurality of objects.

29. The storage architecture of claim 23, wherein said ready access storage area contains temporary processing structures used to organize proposition information by named contextual category, comprising:

a plurality of ordered multi-dimensional context networks, each context network having at least one context dimension, at least one attribute dimension, with at least one tertiary candidate dimension associated with said attribute dimension, said context dimension containing at least one named object representing context;

wherein in any given multi-dimensional context network, when more than one context dimension exists, said dimensions are ordered such that said ordering represents a logical ordering scheme;

wherein at least one attribute dimension is associated with each said context dimension, said attribute dimension containing at least one named attribute object representing a specific attribute applicable to one said named context object of said associated context dimension; and

wherein at least one tertiary candidate dimension is associated with each said attribute dimension, said candidate dimension containing zero, at least one named object, called candidate objects, for each named attribute object in said associated attribute dimension.

30. The storage architecture of claim 23, wherein said architecture further comprises means for adding and removing information from said permanent storage area.

31. The storage architecture of claim 23, wherein said working storage area is divided into a long term memory area and a short term memory area.

32. The storage architecture of claim 23, wherein said working storage area maintains user defaults, parameter information, and adjustable weight information with regard to a predefined subset of propositions.

33. The storage architecture of claim 23, wherein said working storage area maintains a plurality of predefined objects used to facilitate efficient processing.

34. The storage architecture of claim 23, wherein said working storage area maintains a plurality of predefined objects used to establish the structure and initial content of a short term memory area, said short term memory comprising:

at least one predefined list structure, predefined two-dimensional matrix structure, and predefined multi-dimensional network structure, for processing contextual information;

at least one predefined attribute description for each multi-dimensional network structure for processing contextual information;

at least one predefined object to store in said predefined list structure; and

at least one predefined proposition to store in said predefined multi-dimensional network structure for processing contextual information.

35. The storage architecture of claim 23, wherein said permanent storage area maintains information regarding objects, and information regarding propositions formed from a plurality of objects.

36. The storage architecture of claim 29, wherein said candidate objects comprise a proposition formed of a plurality of objects, comprising:

a molecular-style structure having a plurality of objects, wherein said objects function as elements x, y, r, and c and wherein said objects form an explicit proposition comprising at least one of each x, y, r, and c element, wherein x is a first object, y is a second object, r represents a known explicit relationship between said first and second objects, c represents a known logical context of said relationship;

said relationship forming a logical proposition stating a fact regarding said first object; and

said proposition having a predetermined weight assigned thereto; and

said molecular-style structure constructed and arranged such that each said object is capable of being utilized as an element x, y, r, or c with respect to at least one other object.

37. The storage architecture of claim 36, wherein attribute objects are comprised of lexical and non-lexical items and wherein said lexical items are from the group consisting of:

letters, symbols, words, numbers, and combinations thereof.

38. The storage architecture of claim 36, wherein candidate objects are comprised of lexical and non-lexical items and wherein said lexical items are from the group consisting of:

letters, symbols, words, numbers, and combinations thereof.

39. The storage architecture of claim 29, wherein each temporary processing structure contains a header record with at least one information object describing the context of said temporary processing structure and all the information therein.

40. The storage architecture of claim 29, wherein each object in each dimension of each temporary processing structure contains a header record with information objects pertaining to an associated object in said dimension.

41. The storage architecture of claim 29, wherein said proposition information comprises:

the order of objects forming said proposition, the location of said objects in said permanent storage area, a vector for each said proposition comprised of magnitude and direction components, and an adjustable weight for each said proposition.

42. The storage architecture of claim 41, wherein said proposition information further includes:

the original magnitude and direction components for each proposition, and the original weight for each said proposition.

43. The storage architecture of claim 29, wherein a class of objects is represented by a single named object within a dimension and wherein an attribute existing within a class of objects is associated with the dimension containing said named object representing said class.

44. The storage architecture of claim 29, wherein at least one attribute from one of said lower dimensions in said logically ordered scheme may be drawn from one of said higher dimensions in said scheme.

45. The storage architecture of claim 23, wherein said ready access storage area contains temporary processing structures used to organize proposition information by named contextual category, comprising:

a plurality of ordered multi-dimensional context networks, each context network having at least one context dimension as a first dimension, at least one attribute dimension, with an associated tertiary candidate dimension wherein said context dimension containing at least one named object representing context;

wherein in any given multi-dimensional context network, when more than one context dimension exists, said dimensions are ordered such that said ordering represents a logical ordering scheme of super-ordinate and subordinate classes of objects;

wherein at least one attribute dimension is associated with each said context dimension, said attribute dimension containing at least one named object, called an attribute object,

representing a specific attribute applicable to one said named context object of said associated context dimension; and

wherein one tertiary candidate dimension is associated with each said attribute dimension, said candidate dimension containing zero, at least one named object, called candidate objects, for each named attribute object in said associated attribute dimension.

46. A method for the retrieval of data in an object network structure, comprising the steps of;

providing a computer architecture having a object base with a plurality of objects therein, each object representing a letter, number, word, symbol, or data file and each object having a set of propositions linked thereto, said propositions describing the relationships of a first object to at least one other object and said set of propositions accessible by location of said first object;

searching said object base for a predetermined first object; and

accessing said set of propositions.

47. A method for the analysis of the meaning of an ordered group of objects, comprising the steps of;

providing a plurality of propositions formed of two or more objects, each said proposition describing an explicit logical relationship between one object and another object;

comparing all propositions containing a first object of said group of objects to identify any common objects; and

assigning a weight to each said proposition based upon the amount of common objects identified in said proposition.

48. The method of claim 47, further comprising the steps of;

comparing all propositions containing said first object of said group of objects with all other objects of said group of objects to identify any common objects;

assigning a weight to each said proposition based upon the amount of common objects identified in said proposition.

49. The method of claim 47, further comprising the steps of;

comparing all propositions containing a second object of said group of objects to identify any common objects; and

assigning a weight to each said proposition based upon the amount of common objects identified in said proposition.

50. The method of claim 49, further comprising the steps of;

comparing all propositions containing said first object of said group of objects with said propositions containing said second objects of said group of objects to identify any common objects; and

assigning a weight to each said proposition based upon the amount of common objects identified in said proposition.

51. A method for the analysis of the meaning of an ordered group of objects, comprising the steps of;

providing a plurality of propositions ordered sequentially and formed of a plurality of objects, each said proposition describing an explicit logical relationship between one object and another object and having objects representing x, y, r, and c, wherein x is a first object, y is a second object, r is an explicit logical relationship between said first object x and second object y, and c is a logic context in which said logical relationship exists between said first object x and said second object y, said propositions having said objects representing x, y, r, and c ordered respectively;

comparing all propositions containing a first object of said group of objects to identify any common objects;

sequentially examining said propositions for common objects;

extracting y objects from said propositions and reordering them in said proposition as x objects for further sequential examination;

sequentially examining said reordered propositions for common objects;

extracting r relations from said propositions to serve to activate further steps specified by the type or specific identity of the r relation;

sequentially examining said propositions for common objects;

extracting c contexts from said propositions to categorize or group the propositions with other propositions sharing the same or similar c contexts;

sequentially examining said propositions for common objects;

comparing all propositions containing a first object of said group of objects to identify any common objects;

assigning a weight to each said proposition based upon the amount of common objects or the frequency of encountering common objects during sequential examination.

52. A storage architecture for storing digital, analog, or other machine-readable format material, comprising;

an object base having a plurality of objects therein; and

a set of propositions linked to a first object, said set of propositions describing the relationships of a first object to at least one other object, wherein each proposition describes the relationship between said first object and at least one other object.

53. A storage architecture according to claim 52, wherein each said proposition comprises four objects including said first object.

54. A storage architecture according to claim 52, wherein said objects of a given proposition represent the elements x, y, r, and c wherein x is a first object, y is a second object, r represents a known explicit logical relationship, and c represents a known logical context.

55. A storage architecture according to claim 52, wherein said object base is constructed such that said objects are arranged in a single ordered list.

56. A storage architecture according to claim 52, wherein said objects in said object base are divided into at least two groups.

57. A storage architecture according to claim 56, wherein said objects in said object base are divided into groups determined by a predetermined common relationship.

58. A storage architecture according to claim 52, wherein said object base is constructed and arranged as a hash table.

59. The structure according to claim 52, wherein said objects in said object base are comprised of lexical and non-lexical items.

60. A method of for the analysis of the meaning of an ordered group of objects, comprising the steps of;

providing a storage architecture having an object base with a plurality of objects therein;

providing a plurality of propositions formed of a plurality objects, each said proposition describing an explicit logical relationship between one object and another object;

matching a first object, in said group of objects, with an identical object in said object base;

identifying a number of propositions containing said first object; and

examining said propositions for objects common between at least one of said propositions and said ordered group of objects.

61. The method of claim 60, further comprising the steps of:

matching at least one of said common objects with an identical object in said object base;

identifying a second number of propositions containing said at least one common object; and

examining said second number of propositions for objects common between at least two of said second propositions and said ordered group of objects.

62. The method of claim 60, further comprising the steps of:

matching a second object, from said group of objects, with an identical object in said object base;

identifying a second number of propositions containing said second object; and

examining said second propositions for objects common between at least two of said propositions and said ordered objects.

63. The method of claim 60, further comprising the steps of:

examining said propositions having said first object therein and said propositions having said second object therein for objects common between at least two propositions.

64. A method of selecting at least one proposition from a plurality of propositions, the steps of the method comprising:

providing a plurality of propositions, formed from a plurality of objects, each said proposition having a predetermined weight and vector, said weight having a numeric value and said vector having a direction and magnitude;

selecting at least one proposition from said plurality of propositions; and

evaluating said selected propositions based upon said weight values and direction and magnitude of said vectors.

65. The method of claim 64, wherein said vector directions comprise:

rising, static, and falling conditions.

66. The method of claim 64, wherein said vector magnitudes comprise:

above threshold, at threshold, and below threshold conditions.

67. The method of claim 64, wherein said at least one proposition is selected by identifying and selecting propositions having at least one common object forming at least a portion of said proposition.

68. The method of claim 64, wherein said vector magnitude of each proposition is calculated by the number of times said proposition has been correctly selected previously and said direction is calculated from said predetermined weight minus a previous predetermined weight.

69. A method for interpreting text, formed of a plurality of words, comprising the steps of:

providing a lexicon having a plurality of words therein, each said word being linked to a plurality of propositions formed by a plurality of words with each said linked proposition having said words to which said proposition is linked forming a portion of said proposition;

searching for a first word, from said text, in said lexicon;

analyzing the morphology of said word by identifying if a prefix or suffix has been added to a root word to form said first word;

identifying said propositions formed, in part, by said first word;

comparing said identified propositions and selecting propositions having at least one other common word forming at least a portion of at least two said propositions;

searching for a second word of said text in said lexicon;

analyzing the morphology of said word by identifying if a prefix or suffix has been added to a root word to form said second word;

identifying said propositions formed, in part, by said second word;

comparing said identified propositions and selecting propositions having at least one other common word forming at least a portion of at least two of said propositions;

comparing said selected propositions having said first word with said selected propositions having said second word; and

applying a fitness algorithm to determine the most applicable propositions of those compared.

70. The method of claim 69, wherein said fitness algorithm assigns weights to said propositions based upon the number of common words forming at least a portion of at least two of said propositions.

71. The method of claim 69, wherein each said proposition has past weights and a present adjustable weight and past vectors and a present vector having adjustable direction and magnitude attributes assigned thereto, and wherein said fitness algorithm tracks past weights and vectors and determines the fitness of said propositions based upon the analysis of said past and present vectors and weights.

72. The method of claim 69, wherein said fitness algorithm applied to said plurality of propositions comprises formulas that elevate the weight of a proposition having at least one common word forming at least a portion of said proposition when the weight of said proposition is positive.

73. The method of claim 69, wherein said fitness algorithm applied to said plurality of propositions comprises formulas that lower the weight of a proposition having at least one common word forming at least a portion of said proposition when the weight of said proposition is negative.

74. The method of claim 69, wherein said fitness algorithm applied to said plurality of propositions comprises formulas that introduce quasi-random variables to elevate or lower the weight of a proposition.

75. The method of claim 69, wherein said fitness algorithm applied to said plurality of propositions imitates the chaotic flow of electrical potentials in the human brain.

76. A ready access storage area for storing and processing proposition information in digital, analog, and other machine-readable formats, comprising;

at least one context structure comprised of a multi-dimensional network;

at least one structure comprised of a two-dimensional matrix; and

at least one structure comprised of a one-dimensional list of information.

77. A ready access storage area according to claim 76, wherein said context structure contains at least one object organized within said matrices.

78. A ready access storage area according to claim 76, wherein said Matrix structure contains objects that are utilized to define the parameters for management of said context structure.

79. A ready access storage area according to claim 76, wherein said two-dimensional structure provides a list utilized to efficiently access said object within said Matrix and context structures.

80. A method for permitting users to enter language input, comprised of at least one language object, into a language processing system, comprising the steps of:

providing an input mechanism and an output mechanism;

providing at least one input object;

providing a permanent object storage structure having a plurality of objects therein and having information describing the locations of objects having a logic relationship to one another;

matching objects in said permanent object storage structure with objects presented as input objects by individually comparing each of said input objects with an object in permanent storage based on a predetermined similarity factor; and

returning information describing the location of all propositions in permanent storage relating to the matched input object to said output device.

81. The method of claim 80, wherein said at least one input object comprises a sentence.

82. The method of claim 80, wherein said at least one input object comprises a plurality of sentences.

83. The method of claim 80, wherein said at least one input object comprises a string of natural language words and phrases of arbitrary complexity.

84. The method according to claim 80, wherein said input mechanism is from the group consisting of:

keyboards, files, microphones, and speech to text devices.

85. The method according to claim 80, wherein at least one of said input objects is a lexical item.

86. A method for examining the objects of a knowledge molecule having x, r, y, and c objects and a weight, comprising the steps of:

providing a proposition having x, r, y, and c objects and a weight; and

providing a ready access storage area constructed and arranged to have at least one list for the storage of a plurality of objects and at least one weight; and

placing said objects of said proposition in said list in said ready access storage area with said weight.

87. A method according to claim 86, the steps further comprising;

providing said ready access storage area with a multi-dimensional context network, said network having a plurality of objects therein; and

determining if said objects of said proposition match objects in said multi-dimensional context network in said ready access storage area.

88. A method according to claim 87, the steps further comprising;

placing objects of said proposition, from said list into said ready access storage area that match predetermined criteria in a multi-dimensional context network.

89. A method according to claim 88, the steps further comprising:

placing an object of said proposition from said List into a first dimension in said multi-dimensional “

context”

network in said ready access storage area.

90. A method according to claim 89, the steps further comprising inserting an object of said proposition from said List in a secondary dimension to serve as a context object in said secondary dimension of said multi-dimensional context network.

91. A method according to claim 90, the steps further comprising placing an object of said proposition from said List into said secondary dimension in a multi-dimensional context network.

92. A method according to claim 91, the steps further comprising inserting an object of said proposition from said List into said secondary dimension to serve as an attribute in said secondary dimension.

93. A method according to claim 92, the steps further comprising placing an object of said proposition from said List into a tertiary dimension in said multi-dimensional context network.

94. A method according to claim 93, the steps further comprising inserting an object of said proposition from said List into said tertiary dimension to serve as a candidate in said tertiary dimension.

95. A method according to claim 94, the steps further comprising providing an individual weight for each said object in said list;

storing said weight of said object in said List, with said object, in said multi-dimensional context network.

96. A method according to claim 86, wherein said object in said ready access storage is a lexical item.

97. A method according to claim 86, wherein said object in said ready access storage is a knowledge molecule.

98. A method for changing the weight of an object in a multi-dimensional context network in ready access storage, based on a match between one of the objects of a retrieved knowledge molecule and objects in a list in ready access storage, the method comprising:

providing a multi-dimensional network having at least one object stored therein, each said object having a weight assigned thereto, and said network also having at least one knowledge molecule stored therein, each said molecule having a weight assigned thereto;

retrieving said knowledge molecule; and

providing a formula for modifying a weight of said object, said formula for modifying weight of the object governed by said weight of said retrieved knowledge molecule.

99. A method according to claim 98, wherein said formula allows a positive weight of said retrieved knowledge molecule to elevate the weight of said object in said multi-dimensional context network.

100. A method according to claim 98,wherein said formula allows a negative weight of said retrieved knowledge molecule to lower the weight of said object in said multi-dimensional context network.

101. A method according to claim 98,wherein said method further comprises providing a plurality of excitatory and inhibitory modifications to objects in said multi-dimensional context network.

102. A method according to claim 98, the steps further comprising:

introducing random or unrelated variables into said formula used to adjust said weight of said object in said multi-dimensional context network.

103. A method according to claim 102, the steps further comprising monitoring said weight of said object in said ready access storage area to determine if changes in said weight resulting from other processes meets or exceeds a predetermined high or low threshold.

104. A method according to claim 103, the steps further comprising monitoring the change in said weight of said object in said ready access storage area to determine if said change resulting from other processes causes said weight to rise or fall.

105. A method according to claim 104, the steps further comprising recording the rising or falling direction of change in said weight of said object in said ready access storage area.

106. A method according to claim 105, the steps further comprising activating processes based on the magnitude of change of said weight of said object in said ready access storage area.

107. A method according to claim 106, the steps further comprising identifying an object in said ready access storage area whose weight exceeds a predetermined threshold and activating a search in a permanent storage area for knowledge molecules whose objects match the value or values of the weight of said object.

108. A method for preparing an interpreter device to receive natural language input, comprising the steps of:

providing an interpreter device having a working storage area, a ready access storage area and a knowledge network, said ready access storage area having a plurality of objects each having at least one adjustable weight and vector in a context;

retrieving from said working storage area user defaults, parameters, predefined list and context information, words and propositions;

establishing in said ready access storage area reserved storage locations for said predefined lists and contexts;

storing in said predefined lists and contexts in said ready access storage area said predefined words and propositions retrieved from said working storage area; and

initiating simulated electrical potential flow in the knowledge network to adjust the weights of vectors of objects in said contexts in said ready access storage area.

109. A method for preserving context information while an interpreter device is not in operation, comprising the steps of:

providing an interpreter device having a ready access storage area, with context information therein, and a working storage area;

initiating a process, after receiving, but prior to completing a request to shut down an interpreter device, that keeps the device active until the process is complete;

transferring context information from said ready access storage area, to said working storage area for storage in said working storage area;

assigning a calendar date and time to said context information;

storing the calendar date and time applicable to said context information in working storage area with said context information; and

completing said shut down procedure.

110. A method for simulating human consciousness upon activation of an interpreter device, comprising the steps of:

providing an interpreter device having a working storage area, with context information stored therein, and a ready access storage area;

retrieving from said working storage area context information stored in said working storage area prior to shut down; and

storing said context information in said ready access storage area.

111. An interpreter device adapted to automatically obtain location information from a global positioning system constructed and arranged to generate information, comprising;

an interpreter device, having a ready access storage area, said interpreter device being constructed and arranged to receive information from a global positioning system.

112. The device according to claim 111, wherein a global positioning system is attached to the interpreter device as a built-in component.

113. The device according to claim 111, wherein a global positioning system is attached to the interpreter device as an external component.

114. A device according to claim 111, wherein information generated by a global positioning system is retrieved and stored in a predefined context in said ready access storage area of said interpreter device.

115. An interpreter device adapted to automatically obtain date and time information from a time keeping device constructed and arranged to generate information, comprising;

an interpreter device, having a ready access storage area, said interpreter device being constructed and arranged to receive information from a time keeping device.

116. A method of claim 115, wherein information generated by said time keeping device is retrieved and stored in a predefined context in said ready access storage area of said interpreter device.

117. A device for interpreting language, comprising;

an interactive voice response system, said voice response system constructed and arranged to elicit information from a user and respond to a user;

a speech to data interpreter constructed and arranged to translate spoken information into data; and

a multi-dimensional interpreter constructed and arranged to analyze spoken information by analyzing the linguistic features, context, and relationships of the information and generate spoken responses and clarifying questions.

118. A method for interpreting language, comprising the steps of;

providing an interactive voice response system, said voice response system constructed and arranged to elicit information from a user and respond to a user;

providing a speech to data interpreter constructed and arranged to translate spoken information into data objects;

providing a multi-dimensional interpreter constructed and arranged to analyze said data objects by analyzing the linguistic feature, context, and relationships of the data objects;

providing a processor for further processing;

receiving input from a user with said interactive voice response system;

translating said input into data objects with said speech to data interpreter;

interpreting the linguistic features, context, and relationships of said data objects with said multi-dimensional interpreter to identify resulting information for further processing; and

sending the resulting information of the interpretation to a processor for further processing.

119. The method according to claim 118, wherein the processor is constructed and arranged to record said resulting information.

120. The method according to claim 118, wherein the processor is constructed and arranged to place an order based upon the resulting information.

121. The method according to claim 118, wherein the processor is constructed and arranged to provide resulting information to a user via an output device.

122. The method according to claim 118, wherein the processor is constructed and arranged to instruct a device to perform a function based upon directions provided by said resulting information.

123. The method according to claim 118, wherein the step of interpreting further comprises interpreting the context and relationships of said data objects with said multi-dimensional interpreter by analyzing each level of linguistic content of said data objects, wherein the levels include context, semantics, syntax, morphology, phonology, and prosody.

124. The method according to claim 118, wherein the method further comprises the steps of;

initiating said interactive voice response system to ask clarifying questions to a user;

receiving clarifying input from a user with said interactive voice response system;

translating said clarifying input into clarifying data objects with said speech to data interpreter;

interpreting the linguistic features, context, and relationships of said clarifying data objects with said multi-dimensional interpreter to identify clarifying information to supplement said resulting information; and

sending the resulting and clarifying information of the interpretation to a processor for further processing.

125. A storage architecture for storing and processing logically organized information in any machine-readable format, comprising a multi-dimensional context network wherein the network has a plurality of dimensions arranged in a logically ordered scheme and each dimension is directly associated with adjacent dimensions by logical associations, comprising:

a first dimension used to store information representing the most general level of information stored in said multi-dimensional context network;

a second dimension used to store information related to information in said first dimension by a specific association representing a more specific level of information stored in said multi-dimensional context network than said first level;

a tertiary dimension arranged adjacent said second dimension, said tertiary dimension being outside said logically ordered scheme, used to store information related to information in said adjacent second dimension by a specific association representing the same level of specificity of information as in said second dimension;

any number of additional logically ordered dimensions, each of which is used to store information related to information in the dimension immediately above, in said ordered scheme, by a specific association representing a more specific level of information than that within the dimensions above;

at least one additional tertiary dimension adjacent to a logically ordered dimension, said tertiary dimension being outside said logically ordered scheme, each tertiary dimension being used to store information related to information in the adjacent hierarchical dimension by a specific association representing the same level of specificity of information but not part of the hierarchy stored in said multi-dimensional context network.

126. A method of matching each object component of a knowledge molecule with objects in a list in ready access storage and initiating further processing based on each match.

127. A knowledge appliance device with a human language interface designed to interpret human language speech and provide computing and communication services based on commands spoken by a user in a plain human language and interpreted by said device, the device comprising:

at least two interconnected data and instruction transfer busses comprised of at least one primary processing bus, one input bus, and one output bus.

128. The device according to claim 127, wherein said primary processing bus connects at least one main controlling processor unit to said device, said processor unit comprising a general computing services subsystem to coordinate processes and memory shared by other processors.

129. The device according to claim 127, wherein said primary processing bus connects at least one specialized processor unit to said device, said specialized processor unit comprising a speech-to-text and text-to-speech subsystem to convert spoken language user input into digital text and convert digital text into speech.

130. The device according to claim 127, wherein said primary processing bus connects at least one specialized processor unit to said device, said specialized processing unit comprising a language interpretation subsystem to interpret text provided by the speech-to-text analysis subsystem.

131. The device according to claim 127, wherein said primary processing bus connects at least one specialized processor unit to said device, said specialized processor unit constructed and arranged to perform computing services as needed.

132. The device according to claim 127, wherein said primary processing bus connects at least one expansion adaptor subsystem to said device, said expansion adaptor subsystem constructed and arranged to allow at least one additional functional module to be connected to said device.

133. The device according to claim 127, wherein said primary processing bus connects at least one shared Random Access Memory (RAM) to said device, said RAM constructed and arranged to be accessible by any processing units connected to the primary processing bus.

134. The device according to claim 127, wherein said primary processing bus connects at least one non-volatile storage device to said device, said non-volatile storage devices being adapted to support reading, writing, erasure, and modification of data.

135. The device according to claim 127, wherein said primary processing bus connects at least one expansion adaptor to said device, said expansion adaptors constructed and arranged to allow at least one additional storage module to be connected to said device.

136. The device according to claim 127, wherein said primary processing bus connects at least one Read-Only Memory (ROM) module to said device, said ROM module constructed and arranged to contain permanently stored information.

137. The device according to claim 127, wherein a plurality of subsystems are connected to said primary processing hub and wherein said connections between subsystems on said primary processing bus comprise queues and data and instruction paths.

138. The device according to claim 127, wherein said device is connected to at least one specialized processor unit and wherein said specialized processor unit is comprised of commodity microprocessors.

139. The device according to claim 127, wherein said device is connected to at least one specialized processor unit and wherein said specialized processor unit is comprised of at least one reconfigurable microprocessor.

140. The device according to claim 139, wherein said at least one reconfigurable microprocessor is adapted to store part of a device-specific BIOS.

141. The device according to claim 139, wherein said at least one reconfigurable microprocessor is adapted to store part of a device-specific OS.

142. The device according to claim 139, wherein said at least one reconfigurable microprocessor is adapted to store all of a device-specific BIOS.

143. The device according to claim 136, wherein said Read-Only Memory comprised of Erasable and Programmable Read-Only Memory (EPROM).

144. The device according to claim 136, wherein said Read-Only Memory comprised of a non-volatile storage device adapted to enable said appliance device support reading, writing, erasure and modification of data.

145. The device according to claim 136, wherein said Read-Only Memory is comprised of a dedicated portion of a non-volatile storage device adapted to enable said appliance device support reading, writing, erasure and modification of data.

146. The device according to claim 127, wherein said device further comprises at least one memory area and at least one processor, and a device-specific Basic Input/Output System (BIOS), said BIOS adapted to enable said appliance device to coordinate data and instruction flow between processors and memory on said device.

147. The device according to claim 146, wherein said device further comprises non-volatile storage having at least one area and wherein said device-specific Basic Input/Output System (BIOS) adapted to enable said appliance device segregate areas of non-volatile storage for restricted access wherein:

one area of non-volatile storage is dedicated to the BIOS and inaccessible by other applications and processes; and

one area of non-volatile storage is dedicated to an OS and inaccessible by other applications and processes.

148. The device according to claim 127, wherein said input and output bus connects at least one digital signal processor to said device, said digital signal processor adapted to enable said appliance device performs computational operations on signals received by said device.

149. The device according to claim 127, wherein said input and output bus connects at least one cellular phone unit to said device, said cellular phone unit adapted to enable said appliance device to perform wireless communications tasks.

150. The device according to claim 127, wherein said input and output bus connects at least one microphone to said device, said microphone adapted to enable said appliance device to acquire sound waves for use by said device, including sound waves generated by a user speaking into said microphone.

151. The device according to claim 127, wherein said input and output bus connects at least one audio speaker to said device, said speaker adapted to enable said appliance device to produce audio output, including human language word s generated and sequenced by said device.

152. The device according to claim 127, wherein said input and output bus connects at least one video output device to said device, said output device adapted to enable said appliance device to format and transmit, to a video display device, visual output generated by said device.

153. The device according to claim 127, wherein said input and output bus connects at least one printed output device to said device, said printed output device adapted to enable said appliance device to format and transmit, to a printer, visual output generated by said device.

154. The device according to claim 127, wherein said input and output bus connects at least one biometric authentication device to said device, said biometric authentication device adapted to enable said appliance device to receive biometric input from an external biometric scanning device.

155. The device according to claim 127, wherein said input and output bus connects at least one additional input/output adaptor to said device to enable said device to connect other external input and output devices to said device.

156. A method of positively identifying a user, the steps of the method comprising:

placing a predetermined part of a user'"'"'s body in contact with a biometric reader having a scanner;

scanning a body part with the biometric reader and transmitting a digital biometric map of the scanned body part to an input/output subsystem;

transmitting said received digital biometric map from said input/output subsystem to a main processor;

storing the received digital biometric map in a shared memory area of said main processor;

invoking a program within the main processor to retrieve a list of digital biometric maps from a user database stored in an internal storage area of the main processor;

invoking a program within the main processor to compare the received digital biometric map with a plurality of known digital maps in the database; and

storing a name associated with the known digital biometric map in a user identification credential in the shared memory in the main processor if a match is achieved between the received digital biometric map and one of said plurality of known digital maps.

157. A method of positively identifying a user, the steps of the method comprising;

providing a sequence of predefined English words spoken by a user as if spoken to a human being;

providing a device for receiving a stream of sound waves comprising said sequence of words;

providing a microphone attached to said device for receiving said sound waves;

providing an input/output subsystem for transmitting said sound waves to a digital signal processor;

transmitting said received stream of sound waves to the digital signal processor;

creating a sound print of the words spoken by the user with the digital sound processor;

transmitting the received sound print with the input/output subsystem to a main processor;

storing the received sound print in shared memory within the main processor;

invoking a program to retrieve the list of sound prints from the user database stored in the internal storage area within the main processor;

invoking a program to compare the received sound print with a plurality of known sound prints within the database; and

storing the name associated with a known sound print in the within the shared memory area if a match is achieved between the received sound print and a known sound print.

158. A method of providing computing and communication services based on commands spoken by a user in a plain human language, comprising the following steps:

providing a user speaking human language words as if speaking to a human being who understands the human language, thereby creating a stream of sound waves generated by the user;

receiving the stream of sound waves generated by the user through a microphone attached to a device receives;

transmitting said stream of sound waves to a speech-to-text analyzer;

convert said stream of sound waves into a stream of text based on a plurality of speech conversion algorithms;

transmitting said stream of text to the an interpreter;

analyzing analyzes the meaning of said stream of text, by attempting to resolve ambiguity in the words and phrases and infer missing information using a plurality of multi-dimensional interpreter algorithms within an interpreter;

proceeding sequentially through the text and submitting preliminary results to a queue for a device controller subsystem to feed back to said speech-to-text analyzer to establish expectations and to improve the efficiency and accuracy of said speech-to-text analyzer;

generating human language text phrases of pertinent questions intended to complete or improve the interpretation of said stream of text, when said interpreter fails to achieve a sufficient interpretation of said stream of text;

submitting said pertinent questions phrased as text to the text-to-speech system to convert to said text to a spoken language question;

transmitting said spoken language question the speaker to generate output audible to the user;

providing an answer from the user by speaking into said microphone;

searching a list of functions for a function associated with said interpreted text, when said interpreter achieves a sufficient interpretation of said stream of text;

identifying a function matching said interpreted text;

executing program code associated with the matching function, when a matching function is found in said list of functions.

159. A device for processing text by an interpreter comprising:

a currently available computational device; and

an expansion module connected to said computational device, said expansion module used as an accelerator, said expansion module comprising a processor optimally configured to perform processes.

160. A method for expanding and improving the ability of an interpreter to correctly interpret input text, comprising the steps of:

adding a new object to a list of objects in a knowledge network;

adding a new molecular proposition to a network of propositions imitating the human process of learning a new fact;

changing at least one of the objects forming a molecular proposition imitating the human process of learning a new association between objects; and

adjusting a weight associated with a molecular proposition in the network of propositions imitating the human process of learning new information about a known fact.

161. A method of claim 160, wherein an object added to the knowledge network is a lexical object.

162. A method of claim 160, wherein an object added to the knowledge network is a non-lexical object.

163. A method of claim 160, wherein adjusting the weight associated with a molecular proposition in the network of propositions changes the behavior and output of the interpreter.