# US 2013/0159235 A1 · Methods and Systems For Investigation of Compositions of Ontological Subjects > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2013/0159235 A1 - Document ID: US20130159235A1 - Status: Published application - Kind: A1 - Application number: 13/608,333 - Filed: 2012-09-10 - Published: 2013-06-20 - Earliest claimed: 2011-10-11 - Assignees: Hamid Hatami-Hanza - Inventors: Hamid Hatami-Hanza - Web page: https://patentgrep.com/patent/US20130159235A1 ## CPC - G06N 5/04: Inference or reasoning models - G06N 5/02: Knowledge representation; Symbolic representation - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Methods and systems are given for investigation of compositions of ontological subjects in accordance with various aspects of significance. Accordingly, the present invention provide a unified method and process of investigating the compositions of ontological subjects, modeling an unknown system, and obtaining as much worthwhile information and knowledge as possible about the system or the composition or the body of knowledge along with exemplary services utilizing such investigations. ## Claims ### Claim 1 A computer implemented method of investigating compositions of ontological subject utilizing evaluation of one or more of following values for one or more ontological subject of a composition: a. a value respective of at least one of a one or more value significance measures, b. a value respective of at least one of a one or more association strength measures, c. a value respective of at least one of a one or more relative association strength measures, d. a value respective of at least one of a one or more relative value significance measures, e. a value respective of at least one of a one or more novelty value significance measures, f. a value respective of at least one of a one or more relative novelty value significance measures, and g. a value respective of at least one of a one or more association novelty value significance measures. ### Claim 2 (depends on 1) The method of claim 1, wherein one or more of said values are in the form of a vector and/or a matrix. ### Claim 3 (depends on 1) The method of claim 1, wherein the value significance measure of an ontological subject of the composition is a function of frequency of occurrence of the ontological subject in the composition and/or partitions of the composition. ### Claim 4 (depends on 1) The method of claim 1, wherein said association strength measure of a pair of ontological subject of the composition is a function of the co-occurrences of the pair and frequency of occurrences of at least one ontological subject of the pair within the composition or the partitions thereof. ### Claim 5 (depends on 1) The method of claim 1, wherein the composition of ontological subject contains one or more of: a. one or more news content, b. a picture, c. a video, d. a DNA string code, e. a graph and/or a data array representative a graph, f. a computer readable code, g. a multimedia content, h. a textual content, i. a binary string, j. a digital signal, k. an electrical signal, l. a data file, m. one or more web pages, n. a given corpus, and o. a body of knowledge. ### Claim 6 (depends on 1) The method claim 1, wherein the one or more values respective of one or more significance measures of a higher order ontological subject is obtained as a function of one or more value significance measures of its constituent lower order ontological subject. ### Claim 7 A method of investigating a composition of ontological subjects comprising: a. partitioning the composition into a plurality of sets of partitions or ontological subjects each set having labeled with different order, b. evaluating co-occurrences/participation of ontological subjects of at least one predetermined order in at least one of the sets of partitions of another order, c. evaluating value of one or more significance measures based on the participation statistics of ontological subjects in the partitions, using the participation information of a set of lower order OSs into a set of the same or higher order OSs or vice versa, d. processing the ontological subject based on said one or more values. ### Claim 8 (depends on 7) The method of claim 7 further comprises evaluating frequency of occurrences of at least one set of ontological subjects of the composition in a predetermined set of partitions of the composition based on a predetermined definition of frequency of occurrences and/or other than using the data of one or more participation matrices. ### Claim 9 (depends on 7) The method of claim 7 wherein one or more of said significance measures are from intrinsic group of value significances which are functions of values of one or more following measures: a .  vsm_  1  _  1 i k  l = c · f  ( FO i k  l ) b .  vsm_  1  _  2 i k  l = c f  ( FO i k  l ) , i = 1 , 2 , …   N c .  vsm_  2 i k  l = f   ( iop i k  l ) , i = 1   …   N d .  vsm_  3 i k  l = f   ( com ij k  l ) , i = 1   …   N e .  vsm_x i   or   j k  l =  f   ( vsm x i k  l , vsm_y i k  l )   i , j =  1   …   N , x , y =  1 , 2 , … wherein ƒ is a predetermined function in each clause, vsm_xik|l and vsm_yik|l/are data arrays respective of one of a value significance measure, FOik|l is frequency of occurrences, and iopik|l is occurrence probability. ### Claim 10 (depends on 7) The method of claim 7 wherein one or more of significance measures are from associational group of value significances which are functions of values of one or more following measures: a .  asm_  1  _  1 i -> j k  l = com ij k  l   …   i , j = 1   …   N b .  asm_  2  _  1 i -> j k  l =  com ij k  l / vsm_x i k  l   …   i , j =  1   …   N , x =  1 , 2 , … c .  asm_  3  _  1 i -> j k  l =  vsm_y j k  l vsm_x i k  l · com ij k  l   …   i , j =  1   …   N , x , y =  1 , 2 , … d .  asm_  3  _  2 i -> j k  l = c  com ij k  l ( iop i k  l / iop j k  l ) = c  com ij k  l · iop j k  l iop i k  l , i , j = 1   …   N e .  asm_  2  _  2 i -> j k  l = c  com ij k  l iop i k  l , i , j = 1   …   N f .  asm_  4  _  1 i -> j k  l = c · com ij k  l · iop j k  l   i , j = 1   …   N g .  asm_  2  _  3 i -> j k  l = c  ( com ij k  l ) 2 Fo i k  l · Fo j k  l , i , j = 1   …   N h .  asm_x   2 i -> j k  l =  F  ( asm_x   1 i -> j k  l , vsm_x i k , vsm_y j k )   …   i , j =  1   …   N , x , y =  1 , 2 , … , x   1 , x   2 =  1 , 2 , … i .  asm_  3  _  2 i -> j k  l = c · asm_  2  _  2 i -> j k  l · vsm_  1 j k  l j .  ASM_x   3 k  l = ( ASM_x   1 k  l ) T × ASM_x   2 k  l wherein F and c are predetermined functions and/or constants, and/or vectors/matrices, ASMs are association strength/significance measure vectors/matrixes, vsm_xik|l and vsm_yik|l are data arrays respective of one of a value significance measure. ### Claim 11 (depends on 7) The method of claim 7 wherein one or more of significance measures are from relational associational group of value significances which are functions of values of one or more following measures: RASM—1l→k|kl=rasm—1iljkl→k|kl=(PMkl)T×ASMk|lil=1,2, . . . M and jk=1,2, . . . N a. RASM—2l→k|kl=rasm—2iljkl→k|kl=(PMkl)T×ASMk|lTil=1,2, . . . M and jk=1,2, . . . N b. RASM—xk→l|kl=(RASM—xl→k|kl)T c. RASM—xl→l|kl=rasm—xiljll→l|kl=RASM—x1→k|kl×RASM—xl→l|kl,il,jl=1,2, . . . M d. wherein RASMs are relational associational strength/significances and ASMs are associational strength/significance measure vectors/matrixes. ### Claim 12 (depends on 7) The method of claim 7 wherein one or more of significance measures are from relational group of value significances which are functions of values of one or more following measures: rvsm—1—xiljkl→k|kl=(pmikilkl)T×asm—yik→jkk|l . . . ik,jk=1,2, . . . N and il=1,2, . . . M and x,y=1,2, a. RVSM—1—xl→k|kl=RASM—1l→k|kl=rvsm—1iljkl→k|kl=(PMkl)T×ASMk|l,il=1,2, . . . M and jk=1,2, . . . N b. RVSM—2l→k|kl=rvsm—2ilikl→k|kl=(PMkl)T×(ASMk|l)Til=1,2, . . . M and ik=1,2, . . . N c. RVSM—2l→k|kl=RASM—2l→k|kl d. RVSM—3iljkl→k|kl=vsmjkk|l·RASM—11→k|kl=vsmjkk␣l·((PMkl)T×ASMk|l)il=1,2, . . . M and jk=1,2, . . . N e. RVSM—4iljkl→k|kl=vsmjkk|l·RASM—2l→k|kl=vsmjkk|l·((PMkl)T×ASMk|l),il=1,2, . . . M and jk=1,2, . . . N f. RVSM_xiljkl→k|kl=ƒx(vsmjkk|l,RASM—1l→k|kl,RASM—2l→k|kl) g. RVSM—xl→l|kl=rvsm—xiljll→l|kl=RVSM—xl→k|kl×(RVSM_xl→k|kl)T,il,jl=1,2, . . . M h. wherein RVSMs are relational value significances measure vectors/matrixes, vsm_xik|l and asm_yik→jkk|l are data arrays respective of one of a value significance measure and one of association strength/significance measures. ### Claim 13 (depends on 7) The method of claim 7 wherein one or more of significance measures are from relational novel group of value significances which are functions of values of one or more following measures:  a .  rnvsm_  1 i -> j k  l  ( OS i k , OS j k ) ∝ vsm i k  l , vsm j k  l , 1 com ij k  l  b .  rnvsm_  2 i -> j k  l  ( OS i k , OS j k ) ∝ vsm j k  l · 1 com ij k  l  c .  rnvsm_  3 i -> j k  l  ( OS i k , OS j k ) ∝ vsm j k  l , 1 / vsm i k  l , 1 com ij k  l  d .  rnvsm_  4 i -> j k  l  ( OS i k , OS j k ) ∝ 1 / vsm j k  l , 1 / vsm i k  l , com ij k  l  e .  rnvsm_  5 i -> j k  l  ( OS i k , OS j k ) ∝ 1 / vsm j k  l , 1 / vsm i k  l , 1 com ij k  l  f .  rnvsm_  6 i -> j k  l  ( OS i k , OS j k ) ∝ asm i -> j k  l / asm j -> i k  l  g .  rnvsm_  7 i -> j k  l  ( OS i k , OS j k ) ∝ ( asm i -> j k  l asm j -> i k  l + asm j -> i k  l asm i -> j k  l ) h .  rnvsm_x i -> j k  l  ( OS i k , OS j k ) = g 2  ( vsm i k  l , vsm j k  l , asm i -> j k  l , asm j -> i k  l ) , , …   i , j = 1 , 2 , …   N , x = 1 , 2 , …   i .  rnvsm q -> i , j , p k  l  ( OS i k , OS j k , OS p k ) = α 1 · rnvsm_x   1 k  l  ( OS q k , OS i k ) + α 2 · rnvsm_x   2 k  l  ( OS q k , OS j k ) + α 3 · rnvsm_x   3 k  l  ( OS q k , OS p k )     and   q = 1 , 2   …   N j .  rnvsm_x i l ->  j k l -> k  kl = ( pm i k  i l kl ) T × rnvsm_x i k ->  j k k  l   …   i k , j k = 1 , 2 , …   N  and   i l = 1 , 2 , …   M   k .  RNVSM_x l -> k  kl = ( PM kl ) T ×  RNVSM_x k  l   i l = 1 , 2 , …   M  and   j k = 1 , 2 , …   N    l .  RNVSM_x l -> l  kl = RNVSM_x l -> k  kl ×  RNVSM_x k -> l  kl wherein RNVSMs are relational novel value significances measure vectors/matrixes, vsm_xik|l and asm_yik→jkk|l are data arrays respective of one of a value significance measure and one of association strength/significance measures. ### Claim 14 (depends on 8) The method of claim 8 wherein one or more of significance measures are from associational novel group of value significances which are functions of values of one or more following measures: a .  anvsm_  1 i -> j k  l  ( OS i k , OS j k ) ∝ ( asm_x   1 p -> i k  l · asm_x   2 p -> j k  l ) asm_x   3 i -> j k  l ,   p = 1 , 2 , …   N b .  anvsm 2 i -> j k  l  ( OS i k , OS j k ) ∝ ( vsm y   1 i k  l · vsm y   2 j k  l ) × ( asm x   1 p -> i k  l · asm x   2 p -> j k  l ) asm x   3 i -> j k  l ,   p = 1 , 2 , …   N  c .  ANVSM 1 k  l = [ ( ASM x   1 k  l ) T × ASM x   2 k  l ] · ASM x   3 k  l _ d .  anvsm_x i -> j k  l  ( OS i k , OS j k ) = g 3  ( vsm_y   1 i k  l · vsm_y   2 j k  l , asm_x   1 p -> i k  l · asm_x   2 p -> j k  l , asm_x   3 i -> j k  l , asm_x   4 i -> j k  l ) , …   p , i , j = 1 , 2 , …   N , wherein ANVSMs are associational novel value significances measure vectors/matrixes, vsm_xik|l and asm_xi→jk|l are data arrays respective of one of a value significance measure and one of association strength/significance measures. ### Claim 15 (depends on 7) The method of claim 7 wherein one or more of significance measures are from novel group of value significances which are functions of values of one or more following measures: a .  nvsm_  1 i k  l = h 1  ( iop i k  l ) , i = 1 , 2 , …   N b .  nvsm_  1  _  1 i k  l = c / iop i k  l , i = 1 , 2 , …   N c .  nvsm_  1  _  2 i k  l = c / log b  ( iop i k  l ) , i = 1 , 2 , …   N d .  nvsm 1 3 i k  l = c · log b ( 1 iop i k  l ) = - c · log b  ( iop i k  l ) , i = 1 , 2 , …   N  e .  nvsm_  1  _  4 i k  l = - c · log b  ( iop i k  l ) iop i k  l  f .  NVSM_  2 k  l  ( OS i k ) = c  ∑ j   rnvsm_x i -> j k  l  ( OS i k , OS j k )  g .  NVSM_  3 k  l  ( OS j k ) = c  ∑ i   rnvsm_x i -> j k  l  ( OS i k , OS j k )  h .  NVSM_  4 k  l  ( OS i k ) = c  ∑ j   anvsm_x i -> j k  l  ( OS i k , OS j k )  i .  NVSM_  5 k  l  ( OS j k ) = c  ∑ i   anvsm_x i -> j k  l  ( OS i k , OS j k ) j .  NVSM_x k  l  ( OS i k ) = h  ( NVSM_  1 k  l , NVSM_  2 k  l , …   NVSM_y k  l ) , wherein, h1, h and c are predetermined function and/or constants, and/or vectors/matrices, NVSMs are novelty value significance vectors/matrixes, anvsm_xi→jk|l is associational novel value significances, and rnvsm_xi→jk|l are relational novel value significances. ### Claim 16 (depends on 7) The method of claim 7 wherein one or more of significance measures are predetermined vectors or matrices. ### Claim 17 (depends on 7) The method of claim 7 wherein the entries of one or more mathematical objects are transformed in such a way to obtain one or more mathematical objects with one or more desired characteristics. ### Claim 18 The method of clam 7 wherein the composition of ontological subjects contains one or more of followings: a. one or more news content, b. a picture, c. a video, d. a DNA string code, e. a graph and/or a data array representative a graph, f. a computer readable code, g. a multimedia content, h. a textual content, i. a binary string, j. a digital signal, k. an electrical signal, l. one or more web pages, m. a given corpus, n. a body of knowledge, and o. a data file. ### Claim 19 A system of investigating compositions of ontological subjects comprising: a. facilities to receive an input electronic contents via a communication networks wherein the input content can include different compositions of ontological subjects and/or queries and/or controlling instruction and/or indications of selection of one or more service options, b. a computer implemented module to generate or access one or more data arrays respective of one or more participation matrices, c. computer executable instructions that when executed by one or more computer systems, the computer system having at least one processing and/or at least one storage media device, cause the one or more computers to evaluate one or more values respective of one or more significance measures of one or more constituent ontological subjects of the composition whereby a desired response is selected and/or composed in response of the received electronic content, d. facilities to store and/or communicate and/or sending the desired response to one or more agent clients or one or more users. ### Claim 20 (depends on 19) The system of claim 19, wherein the composition of ontological subjects contains one or more of: a. one or more news content, b. a picture, c. a video, d. a DNA string code, e. a graph and/or a data array representative a graph, f. a computer readable code, g. a multimedia content, h. a textual content, i. a binary string, j. a digital signal, k. an electrical signal, l. a data file, m. one or more web pages, n. a given corpus, and o. a body of knowledge.