Artificial intelligence method and apparatus

• US 8,700,620 B1
• Filed: 04/26/2011
• Issued: 04/15/2014
• Est. Priority Date: 04/27/2010
• Status: Active Grant

• Current Assignee: Jeremy Lieberman
• Original Assignee: Jeremy Lieberman
• Inventors: Lieberman, Jeremy
• Primary Examiner(s): Smith, Brannon W
• Application Number:
  US13/094,793
• Time in Patent Office: 1,085 Days
• Field of Search: 707/706, 707/999.005, 707/728
• US Class Current: 707/728
• CPC Class Codes: G06F 16/36 Creation of semantic tools, e.g. ontology or thesauri

Abstract

The invention entails a method of Etiological prioritization based on an original symbolic model of the causative construction of singular logical relations which are built from an initiation of sensed masses in patterns, the naming of sensed patterns into words, the description of the relations of masses using specific English language primitives, the relations between masses and words, and the relations between primitive words. This process is built upon a theory and model of discretized human capabilities and learned experiential meanings. Human capabilities and learned experiential meanings are identified as singular primitives and amount to words identified and matched to singular parts that make up human capabilities and the inventive model of the experiences of the sensing and understanding of the spatiotemporal physical world.

Claims

1. A method, comprising:

(a) creating a first database of meaning items, each of said meaning items being associated with a particular meaning, said first database being created by;

(i) identifying a plurality of primitive words in a first group, said primitive words in said first group having a singular meaning;

(ii) storing said plurality of primitive words in said first group having a singular meaning at a first set of storage locations in a computing device;

(iii) identifying a plurality of primitive words in a second group, said plurality of primitive words in said second group having an associated meaning or meanings which may be defined by said primitive words in said first group and/or other primitive words in said second group;

(iv) storing said plurality of primitive words in said second group at a second set of storage locations in a computing device;

(v) for each of said primitive words in said second group, storing its respective associated meanings as definitional linking information, linking a respective primitive word in said second group to respective defining primitive words in said first group and/or other primitive words in said second group, whereby upon the identification of a word, a definition comprising linked primitive words may be retrieved;

(vi) for each of a plurality of definable words, wherein said definable words are words in said first and second groups,

(A) selecting words from said first and second groups as relative words having the same meaning as a respective definable word, and/or

(B) associating words in said first and second groups into respective relative phrases having the same meaning as a respective definable word, each of said definable words, together with its respective relative words and relative phrases forming a relative set; and

(vii) for each of said definable words, storing relative set linking information linking each of said definable words to its respective relative words and relative phrases, whereby upon the identification of a word, the other members of a relative set may be retrieved;

(b) receiving a query;

(c) comparing the words in the query to said meaning items to determine those words which are not included within said first database of meaning items, said words of said query which are not included within said first database of meaning items forming a keyword set comprising jargon-type words, and the words in said query found in said first database of meaning items forming a meaning search set comprising non-jargon meaning words;

(d) inputting said keyword set into a search engine;

(e) receiving a plurality of documents from said search engine;

(f) determining whether said meaning items in said meaning search set are in each of said documents;

(g) tallying, in a meaning database, occurrences of said meaning items in said documents;

(h) ranking the documents based upon said occurrences tallied in said meaning database to select a plurality of top-ranked documents; and

(i) identifying said top-ranked documents.

2. A method as in claim 1, wherein said determining whether said meaning items in said meaning search set are in each of said documents is done by searching for occurrences of said relative sets of words and/or statements with equivalent meaning.

3. A method as in claim 1, wherein said keyword searching is done by an enterprise search engine or a search engine accessing a public network such as the Internet.

4. The method of claim 1, wherein the database is not dictionary derived, encyclopedia derived, or derived from information in a database derived using another flexibly-ruled or unstructured method.

5. A method as in claim 1, wherein the database includes statements based on the encoding and etiological prioritization of procedures, properties, rules, and definitions from English and/or specialized domains.

6. A method as in claim 1, wherein the smallest number of primitives are combined with one another to create a singular context, property, procedure, or definition within a domain to create higher level terms in a prioritized data set.

7. A method as in claim 1, wherein a portion of said primitive words in said first group are derived through an etiological logical progression of primitives, said primitives being progressively built using a model of causation, whereby basic singular forms, starting from a sensing of patterns of things, are built, based on describing the relationships between multiple patterns of said things, said relationships being words in said first group;

using the relationships amongst these sensed things to both build and assign meaning to words in said first and second groups to form an ordered hierarchy of interrelationships, said words in said first and second groups being combined with things and other things or words in said first and second groups to build the meanings of new words in said first and second groups, the continued development of relationships between primitives and things assigning meaning to and/or building etiologically prioritized meanings of human based capabilities and experiences, such as recognition, generation, action, want, belief, feelings, new meanings being built via the combination of singular primitives representing human capabilities, such as senses, recognitions, feelings, belief, generations, combined with words in said first and second groups not representing human capabilities, such as things, operators, prepositions, adjectives, pronouns, interjections, verbs, nouns, and probability.

8. A method, comprising:

(a) identifying a plurality of primitive words and phrases in a first group, said primitive words and phrases in said first group having a singular meaning;

(b) storing said plurality of primitive words and phrases in said first group having a singular meaning at a first set of storage locations in a computing device;

(c) identifying a plurality of primitive words and phrases in a second group, said plurality of primitive words and phrases in said second group having an associated meaning or meanings which may be defined by said primitive words in said first group and/or other primitive words and phrases in said second group;

(d) storing said plurality of primitive words in said second group at a second set of storage locations in a computing device;

(e) for each of said primitive words in said second group, storing its respective associated meanings as definitional linking information, linking a respective primitive word in said second group to respective defining primitive words in said first group and/or other primitive words in said second group, whereby upon the identification of any word and phrase, a definition comprising linked primitive words may be retrieved;

(f) for each of a plurality of definable words, wherein said definable words are words and phrases in said first and second groups,

(i) selecting words and phrases from said first and second groups as relative words having the same meaning as a respective definable word, and/or

(ii) associating words and phrases in said first and second groups into respective relative phrases having the same meaning as a respective definable word, each of said definable words, together with its respective relative words and relative phrases forming a relative set;

(g) for each of said definable words, storing relative set linking information linking each of said definable words and phrases to their respective relative words and relative phrases, whereby upon the identification of any word, the other members of a relative set may be retrieved;

(h) receiving a query;

(i) comparing the words in the query to said words in said first and second groups to determine those words which are not included within said first and second groups said not included words forming a keyword search set, the words in said query found in said first and second groups forming a meaning search set;

(j) inputting said words in said meaning search set into a search engine, said search engine accessing a database of documents;

(k) determining whether said words in said meaning search set are in each of said documents;

(l) tallying in a search results database occurrences in said documents of said words in said meaning search set;

(m) ranking the documents based upon said occurrences tallied in said search results database to select a plurality of top-ranked documents;

(n) inputting said keyword set into said search engine;

(o) determining whether the words in said keyword set are in each of said documents;

(p) tallying in said search results database occurrences in said documents of said words in said keyword set, wherein said ranking of said documents is also based upon said occurrences of said words in said keyword set tallied in said search results database; and

(q) identifying said top-ranked documents.

9. A method as in claim 8, wherein said second group is divided into a combinatorial group and a terminal group, said combinatorial group comprising words having singular and multiple meaning components, and said terminal group comprising words having multiple meaning components.

10. A method as in claim 8, wherein said primitive words in said first group are generated by a method comprising postulating a sensing of a pattern of things;

recognizing said sensed thing;

requiring a word to represent a thing, requiring a word to represent a word, and using a combination of things and words to make more definitionally complex words.

11. A method as in claim 8, wherein a portion of said primitive words in said first group are derived through an etiological logical progression of primitive words, said primitives being progressively built using a model of causation, whereby basic singular forms, starting from a sensing of patterns of things, are built, based on describing the relationships between multiple patterns of said things, said relationships being words in said first group;

using the relationships amongst these sensed things to both build and assign meaning to words in said first and second groups to form an ordered hierarchy of interrelationships, said words in said first and second groups being combined with things and other things or words in said first and second groups to build the meanings of new words in said first and second groups, the continued development of relationships between primitives and things assigning meaning to and/or building etiologically prioritized meanings of human based capabilities and experiences, such as recognition, generation, action, want, belief, feelings, new meanings being built via the combination of singular primitives representing human capabilities, such as senses, recognitions, feelings, belief, generations, combined with words in said first and second groups not representing human capabilities, such as things, operators, prepositions, adjectives, pronouns, interjections, verbs, nouns, and probability.

12. A method as in claim 8 wherein said first group has forty or more words selected from the group consisting of the words mass, thought, emotion, feeling, sense, recognition, generation, action, truth, place, positive, negative, neutral, not, is, personalization, stop, continue, start, question, answer, amount, name, again, gain, loss, displace, move, change, time, like, dislike, and, attention, not, direction, a, the, for, because, action/do, real, position, proximity, from, to, forward, backward, side, energy, in, out, on, below, over, under, sad, happy, anger, nervous, fear, good, bad, away, smile, frown, land, know, same, different, melancholy, notice, say, read, write, and curious, or words which are equivalent thereto.

13. A method as in claim 8, wherein said identifying the plurality of primitive words having a singular meaning in the first group comprises asking about unknown or changing items.

14. A method as in claim 8, wherein said primitive words in said first and second groups form a prioritized data set.

15. A method as in claim 8, wherein a series of questions and answers is used to ask about unknown or changing items to query about unknown actions and events by adjoining to augment search queries and resulting results.

16. A method as in claim 8, wherein the smallest number of primitives are combined with one another to create a singular context, property, procedure, or definition within a domain to create higher level terms in a prioritized data set.

17. A method as in claim 8, wherein the smallest number of primitives are combined with one another to create a singular context, property, procedure, or definition within a domain to create higher level terms in a prioritized data set.

18. A method comprising:

(a) identifying a plurality of primitive words and phrases in a first group, said primitive words and phrases in said first group having a singular meaning;

(b) storing said plurality of primitive words and phrases in said first group having a singular meaning at a first set of storage locations in a computing device;

(c) identifying a plurality of primitive words and phrases in a second group, said plurality of primitive words and phrases in said second group having an associated meaning or meanings which may be defined by said primitive words in said first group and/or other primitive words and phrases in said second group;

(d) storing said plurality of primitive words in said second group at a second set of storage locations in a computing device;

(e) for each of said primitive words in said second group, storing its respective associated meanings as definitional linking information, linking a respective primitive word in said second group to respective defining primitive words in said first group and/or other primitive words in said second group, whereby upon the identification of any word and phrase, a definition comprising linked primitive words may be retrieved;

(f) for each of a plurality of definable words, wherein said definable words are words and phrases in said first and second groups,

(i) selecting words and phrases from said first and second groups as relative words having the same meaning as a respective definable word, and/or

(ii) associating words and phrases in said first and second groups into respective relative phrases having the same meaning as a respective definable word, each of said definable words, together with its respective relative words and relative phrases forming a relative set;

(g) for each of said definable words, storing relative set linking information linking each of said definable words and phrases to their respective relative words and relative phrases, whereby upon the identification of any word, the other members of a relative set may be retrieved;

(h) receiving a query;

(i) comparing the words in the query to said words and phrases in said first and second groups to determine those words which are not included within said first and second groups said not included words forming a keyword search set, the words in said query found in said first and second groups forming a meaning search set;

(j) inputting into a search engine a group of words and phrases, comprising;

i. said words and phrases in said meaning search set,

ii. the words in the definitions of said words and phrases in said meaning search set, and

iii. the relative sets of said words and phrases in said meaning search set, and the relative sets of said words and phrases in the definitions of said words in said meaning search set;

(k) determining whether words and phrases from said group of words and phrases are in each of a plurality of documents being accessed by said search engine;

(l) tallying in a search results database occurrences in said documents of said words and phrases from said group of words and phrases;

(m) ranking the documents based upon said occurrences tallied in said search results database to select a plurality of top-ranked documents;

(n) identifying said top-ranked documents;

(o) inputting said keyword set into said search engine;

(p) determining whether the words in said keyword set are in each of said documents; and

(q) tallying in said search results database occurrences in said documents of said words in said keyword set, wherein said ranking of said documents is also based upon said occurrences of said words in said keyword set tallied in said search results database.

19. A method as in claim 18, wherein said determining is done on a paragraph by paragraph basis, and said ranking is done by ranking individual paragraphs, and then assigning document ranks on the basis of their constituent paragraph rankings.

20. A method as in claim 19, wherein said top-ranked documents are identified by outputting top ranked paragraphs and hyperlinking to their respective documents.

21. A method as in claim 18, wherein a portion of said primitive words in said first group are derived through an etiological logical progression of primitives, said primitives being progressively built using a model of causation, whereby basic singular forms, starting from a sensing of patterns of things, are built, based on describing the relationships between multiple patterns of said things, said relationships being words in said first group;

using the relationships amongst these sensed things to both build and assign meaning to words in said first and second groups to form an ordered hierarchy of interrelationships, said words in said first and second groups being combined with things and other things or words in said first and second groups to build the meanings of new words in said first and second groups, the continued development of relationships between primitives and things assigning meaning to and/or building etiologically prioritized meanings of human based capabilities and experiences, such as recognition, generation, action, want, belief, feelings, new meanings being built via the combination of singular primitives representing human capabilities, such as senses, recognitions, feelings, belief, generations, combined with words in said first and second groups not representing human capabilities, such as things, operators, prepositions, adjectives, pronouns, interjections, verbs, nouns, and probability.