METHODS AND SYSTEMS FOR KNOWLEDGE DISCOVERY

• US 20100049684A1
• Filed: 10/15/2007
• Published: 02/25/2010
• Est. Priority Date: 10/13/2006
• Status: Active Grant

• Current Assignee: Elsevier Incorporated (RELX PLC)
• Original Assignee: Elsevier Incorporated (RELX PLC)
• Inventors: Adriaansen, Edwin, Schijvenaars, Bob
• Granted Patent: US 8,600,922 B2
• Time in Patent Office: Days
• US Class Current: 706/46
• CPC Class Codes: G06F 16/245 Query processing
• CPC Class Codes: G06F 40/117 Tagging; Marking up details of markup languages G06F40/143; Designating a block; Setting of attributes style sheets, e.g. eXtensible Stylesheet Language Transformation [XSLT], G06F40/154
• CPC Class Codes: G06F 40/237 Lexical tools
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06T 11/001 Texturing; Colouring; Generation of texture or colour inpainting G06T5/77
• CPC Class Codes: G06T 11/20 Drawing from basic elements, e.g. lines or circles

Abstract

Provided are methods and systems for knowledge discovery utilizing knowledge profiles.

Claims

1. A method for textual analysis comprising:

a. determining a co-occurrence of a long form and an associated short form of a term in a document;

b. locating a plurality of occurrences of the associated short form; and

c. expanding the plurality of occurrences of the associated short form with the long form wherein the document has a more accurate representation of frequency of occurrence of the term.

2. The method of claim 1, wherein the long form of the term comprises at least one word.

3. The method of claim 2, wherein the associated short form is an abbreviation of the at least one word.

4. The method of claim 1, wherein the term represents a concept.

5. The method of claim 1, further comprising determining a frequency of occurrence of the term in the document.

6. The method of claim 1, further comprising generating a fingerprint of the document.

7. The method of claim 1, further comprising performing steps a-c for a plurality of documents.

8. A system for textual analysis comprising:

a memory configured for storing text data; and

a processor, coupled to the memory, configured for performing steps comprising,

a. determining a co-occurrence of a long form and an associated short form of a term in a document,

b. locating a plurality of occurrences of the associated short form, and

c. expanding the plurality of occurrences of the associated short form with the long form wherein the document has a more accurate representation of frequency of occurrence of the term.

9. The system of claim 8, wherein the long form of the term comprises at least one word.

10. The system of claim 9, wherein the associated short form is an abbreviation of the at least one word.

11. The system of claim 8, wherein the term represents a concept.

12. The system of claim 8, wherein the processor is further configured for determining a frequency of occurrence of the term in the document.

13. The system of claim 8, wherein the processor is further configured for generating a fingerprint of the document.

14. The system of claim 8, wherein the processor is further configured for performing steps a-c for a plurality of documents.

15. A computer readable medium with computer executable instructions embodied thereon for textual analysis comprising:

a. determining a co-occurrence of a long form and an associated short form of a term in a document;

b. locating a plurality of occurrences of the associated short form; and

c. expanding the plurality of occurrences of the associated short form with the long form wherein the document has a more accurate representation of frequency of occurrence of the term.

16. The computer readable medium of claim 15, wherein the long form of the term comprises at least one word.

17. The computer readable medium of claim 16, wherein the associated short form is an abbreviation of the at least one word.

18. The computer readable medium of claim 15, wherein the term represents a concept.

19. The computer readable medium of claim 15, further comprising computer executable instructions for determining a frequency of occurrence of the term in the document.

20. The computer readable medium of claim 15, further comprising computer executable instructions for generating a fingerprint of the document.

21. The computer readable medium of claim 15, further comprising computer executable instructions for performing steps a-c for a plurality of documents.

22. A method for textual analysis comprising:

receiving a fingerprint for each of a plurality of documents; and

combining fingerprints having a relationship.

23. The method of claim 22, wherein the fingerprint is a list of concepts and their associated relevance weights.

24. The method of claim 22, wherein combining fingerprints having a relationship comprises averaging the fingerprints.

25. The method of claim 23, wherein combining fingerprints having a relationship comprises:

taking a square of the associated relevance weights;

averaging the squares of the weights, and

taking the root of the averages.

26. The method of claim 22, wherein the relationship comprises at least one of, stems from the same author, published in the same publication, contains the same concept, or stems from the same organization.

27. A system for textual analysis comprising:

a memory configured for storing text data; and

a processor, coupled to the memory, configured for performing steps comprising,

receiving a fingerprint for each of a plurality of documents; and

combining fingerprints having a relationship.

28. The system of claim 27, wherein the fingerprint is a list of concepts and their associated relevance weights.

29. The system of claim 27, wherein combining fingerprints having a relationship comprises averaging the fingerprints.

30. The system of claim 28, wherein combining fingerprints having a relationship comprises:

taking a square of the associated relevance weights;

averaging the squares of the weights, and

taking the root of the averages.

31. The system of claim 27, wherein the relationship comprises at least one of, stems from the same author, published in the same publication, contains the same concept, or stems from the same organization.

32. A computer readable medium with computer executable instructions embodied thereon for textual analysis comprising:

receiving a fingerprint for each of a plurality of documents; and

combining fingerprints having a relationship.

33. The computer readable medium of claim 32, wherein the fingerprint is a list of concepts and their associated relevance weights.

34. The computer readable medium of claim 32, wherein combining fingerprints having a relationship comprises averaging the fingerprints.

35. The computer readable medium of claim 33, wherein combining fingerprints having a relationship comprises:

taking a square of the associated relevance weights;

averaging the squares of the weights, and

taking the root of the averages.

36. The computer readable medium of claim 32, wherein the relationship comprises at least one of, stems from the same author, published in the same publication, contains the same concept, or stems from the same organization.

37. A method for textual analysis comprising:

generating a fingerprint for each of a plurality of documents; and

combining fingerprints having a relationship.

38. The method of claim 37, wherein the fingerprint is a list of concepts and their associated relevance weights

39. The method of claim 37, wherein combining fingerprints having a relationship comprises averaging the fingerprints.

40. The method of claim 38, wherein combining fingerprints having a relationship comprises:

taking a square of the associated relevance weights;

averaging the squares of the weights, and

taking the root of the averages.

41. The method of claim 37, wherein the relationship is at least one of, same author, same publication, same concept, and same institution.

42. A system for textual analysis comprising:

a memory configured for storing text data; and

a processor, coupled to the memory, configured for performing steps comprising,

generating a fingerprint for each of a plurality of documents; and

combining fingerprints having a relationship.

43. The system of claim 42, wherein the fingerprint is a list of concepts and their associated relevance weights

44. The system of claim 42, wherein combining fingerprints having a relationship comprises averaging the fingerprints.

45. The system of claim 43, wherein combining fingerprints having a relationship comprises:

taking a square of the associated relevance weights;

averaging the squares of the weights, and

taking the root of the averages.

46. The system of claim 42, wherein the relationship is at least one of, same author, same publication, same concept, and same institution.

47. A computer readable medium with computer executable instructions embodied thereon for textual analysis comprising:

generating a fingerprint for each of a plurality of documents; and

combining fingerprints having a relationship.

48. The computer readable medium of claim 47, wherein the fingerprint is a list of concepts and their associated relevance weights

49. The computer readable medium of claim 47, wherein combining fingerprints having a relationship comprises averaging the fingerprints.

50. The computer readable medium of claim 48, wherein combining fingerprints having a relationship comprises:

taking a square of the associated relevance weights;

averaging the squares of the weights, and

taking the root of the averages.

51. The computer readable medium of claim 47, wherein the relationship is at least one of, same author, same publication, same concept, and same institution.

52. A method for predicting a potential relationship between concepts without a co-occurrence comprising:

receiving a context fingerprint for each of a plurality of concepts;

determining an overlap of context fingerprints among the plurality of concepts; and

predicting that two or more of the plurality of concepts have a relationship wherein the overlap is above a first threshold.

53. A method for predicting a potential relationship between concepts without a co-occurrence comprising:

receiving a context fingerprint for each of a plurality of concepts;

determining a similarity score between the plurality of concepts; and

predicting that two or more of the plurality of concepts have a relationship wherein the similarity score is above a first threshold.

54. A method for predicting a potential relationship between concepts without a co-occurrence comprising:

receiving a context fingerprint for each of a plurality of concepts;

determining an overlap of context fingerprints among the plurality of concepts;

determining a similarity score between the context fingerprints; and

predicting that two or more of the plurality of concepts have a relationship wherein the overlap is above a first threshold and the similarity score is above a second threshold.

55. The method of claim 54, wherein the plurality of concepts do not co-occur in a plurality of documents.

56. The method of claim 54, wherein the plurality of concepts do not co-occur within the same sentence of a single document.

57. The method of claim 54, wherein the plurality of concepts do not co-occur within the same paragraph of a single document.

58. The method of claim 54, wherein a context fingerprint is a list of concepts and their associated relevance weights which is constructed based on co-occurrence of concepts in documents with the concept the context fingerprint is created for.

59. The method of claim 54, wherein determining an overlap of context fingerprints among the plurality of concepts comprises determining a number of concepts the two context fingerprints have in common.

60. The method of claim 54, wherein determining a similarity score between the context fingerprints comprises performing a matching algorithm.

61. The method of claim 60, wherein performing a matching algorithm comprises:

storing each context fingerprint as a vector; and

performing a vector matching algorithm.

62. A system for predicting a potential relationship between concepts without a co-occurrence comprising:

a memory configured for storing textual data;

a processor, coupled to the memory, configured for performing steps comprising,

receiving a context fingerprint for each of a plurality of concepts;

determining an overlap of context fingerprints among the plurality of concepts;

determining a similarity score between the context fingerprints; and

predicting that two or more of the plurality of concepts have a relationship wherein the overlap is above a first threshold and the similarity score is above a second threshold.

63. The system of claim 62, wherein the plurality of concepts do not co-occur in a plurality of documents.

64. The system of claim 62, wherein the plurality of concepts do not co-occur within the same sentence of a single document.

65. The method of claim 62, wherein the plurality of concepts do not co-occur within the same paragraph of a single document.

66. The system of claim 62, wherein a context fingerprint is a list of concepts and their associated relevance weights which is constructed based on co-occurrence of concepts in documents with the concept the context fingerprint is created for.

67. The system of claim 62, wherein determining an overlap of context fingerprints among the plurality of concepts comprises determining a number of concepts the two context fingerprints have in common.

68. The system of claim 62, wherein determining a similarity score between the context fingerprints comprises performing a matching algorithm.

69. The system of claim 68, wherein performing a matching algorithm comprises:

storing each context fingerprint as a vector; and

performing a vector matching algorithm.

70. A computer readable medium with computer executable instructions embodied thereon for predicting a potential relationship between concepts without a co-occurrence comprising:

receiving a context fingerprint for each of a plurality of concepts;

determining an overlap of context fingerprints among the plurality of concepts;

determining a similarity score between the context fingerprints; and

predicting that two or more of the plurality of concepts have a relationship wherein the overlap is above a first threshold and the similarity score is above a second threshold.

71. The computer readable medium of claim 70, wherein the plurality of concepts do not co-occur in a plurality of documents.

72. The computer readable medium of claim 70, wherein the plurality of concepts do not co-occur within the same sentence of a single document.

73. The computer readable medium of claim 70, wherein the plurality of concepts do not co-occur within the same paragraph of a single document.

74. The computer readable medium of claim 70, wherein a context fingerprint is a list of concepts and their associated relevance weights which is constructed based on co-occurrence of concepts in documents with the concept the context fingerprint is created for.

75. The computer readable medium of claim 70, wherein determining an overlap of context fingerprints among the plurality of concepts comprises determining a number of concepts the two context fingerprints have in common.

76. The computer readable medium of claim 70, wherein determining a similarity score between the context fingerprints comprises performing a matching algorithm.

77. The computer readable medium of claim 76, wherein performing a matching algorithm comprises:

storing each context fingerprint as a vector; and

performing a vector matching algorithm.

78. A method for editing a fingerprint comprising:

presenting, to a user, at least one concept that comprises a fingerprint;

presenting, to the user, a user interface wherein the user can adjust an importance of the at least one concept, wherein the importance is a weight that is adjustable from −

1 to +1 or the importance is an attribute expressing that the concept is forbidden or required;

receiving an adjusted importance from the user; and

applying the adjusted importance to the concept.

79. The method of claim 78, wherein presenting, to a user, at least one concept that comprises a fingerprint comprises displaying a list of concepts to the user on a display device.

80. The method of claim 78, wherein the user interface comprises a sliding scale wherein a user can drag a point along a line from −

1 to +1.

81. The method of claim 78, wherein the user interface comprises a plurality of inputs wherein each input is sequentially numbered from −

1 to +1 and a user can select an input.

82. The method of claim 78, wherein the user interface comprises a first and a second input wherein the first input corresponds to forbidding the concept and the second input corresponds to requiring the concept and a user can select one of the first or second inputs.

83. The method of claim 78, further comprising performing a search with the fingerprint comprising adjusted weights as a query.

84. The method of claim 78, further comprising performing a search with a concept that is required.

85. The method of claim 78, further comprising performing a search with a concept that is forbidden.

86. A system for editing a fingerprint comprising:

a memory configured for storing textual data;

a processor, coupled to the memory, configured for performing steps comprising,

presenting, to a user, at least one concept that comprises a fingerprint;

presenting, to the user, a user interface wherein the user can adjust an importance of the at least one concept, wherein the importance is a weight that is adjustable from −

1 to +1 or the importance is an attribute expressing that the concept is forbidden or required;

receiving an adjusted importance from the user; and

applying the adjusted importance to the concept.

87. The system of claim 86, wherein presenting, to a user, at least one concept that comprises a fingerprint comprises displaying a list of concepts to the user on a display device.

88. The system of claim 86, wherein the user interface comprises a sliding scale wherein a user can drag a point along a line from −

1 to +1.

89. The system of claim 86, wherein the user interface comprises a plurality of inputs wherein each input is sequentially numbered from −

1 to +1 and a user can select an input.

90. The system of claim 86, wherein the user interface comprises a first and a second input wherein the first input corresponds to forbidding the concept and the second input corresponds to requiring the concept and a user can select one of the first or second inputs.

91. The system of claim 86, further comprising performing a search with the fingerprint comprising adjusted weights as a query.

92. The system of claim 86, further comprising performing a search with a concept that is required.

93. The system of claim 86, further comprising performing a search with a concept that is forbidden.

94. A computer readable medium with computer executable instructions embodied thereon for editing a fingerprint comprising:

presenting, to a user, at least one concept that comprises a fingerprint;

presenting, to the user, a user interface wherein the user can adjust an importance of the at least one concept, wherein the importance is a weight that is adjustable from −

1 to +1 or the importance is an attribute expressing that the concept is forbidden or required;

receiving an adjusted importance from the user; and

applying the adjusted importance to the concept.

95. The computer readable medium of claim 94, wherein presenting, to a user, at least one concept that comprises a fingerprint comprises displaying a list of concepts to the user on a display device.

96. The computer readable medium of claim 94, wherein the user interface comprises a sliding scale wherein a user can drag a point along a line from −

1 to +1.

97. The computer readable medium of claim 94, wherein the user interface comprises a plurality of inputs wherein each input is sequentially numbered from −

1 to +1 and a user can select an input.

98. The computer readable medium of claim 94, wherein the user interface comprises a first and a second input wherein the first input corresponds to forbidding the concept and the second input corresponds to requiring the concept and a user can select one of the first or second inputs.

99. The computer readable medium of claim 94, further comprising performing a search with the fingerprint comprising adjusted weights as a query.

100. The computer readable medium of claim 94, further comprising performing a search with a concept that is required.

101. The computer readable medium of claim 94, further comprising performing a search with a concept that is forbidden.

102. A method for re-generating a fingerprint comprising:

identifying a concept and a synonym that have been added to a terminology system;

identifying a document that is affected wherein identifying comprises searching a word based index; and

re-generating a fingerprint for the affected document.

103. The method of claim 102, wherein the terminology system is a thesaurus.

104. The method of claim 102, wherein the word based index comprises a list of words and documents containing each word.

105. A system for re-generating a fingerprint comprising:

a memory configured for storing textual data;

a processor, coupled to the memory, configured for performing steps comprising,

identifying a concept and a synonym that have been added to a terminology system;

identifying a document that is affected wherein identifying comprises searching a word based index; and

re-generating a fingerprint for the affected document.

106. The system of claim 105, wherein the terminology system is a thesaurus.

107. The system of claim 105, wherein the word based index comprises a list of words and documents containing each word.

108. A computer readable medium with computer executable instructions embodied thereon for re-generating a fingerprint comprising:

identifying a concept and a synonym that have been added to a terminology system;

identifying a document that is affected wherein identifying comprises searching a word based index; and

re-generating a fingerprint for the affected document.

109. The computer readable medium of claim 108, wherein the terminology system is a thesaurus.

110. The computer readable medium of claim 108, wherein the word based index comprises a list of words and documents containing each word.

111. A method for knowledge discovery comprising:

receiving a query;

generating a fingerprint for the query, wherein the fingerprint comprises at least one concept;

determining at least one related concept to the fingerprint according to a terminology system; and

expanding the fingerprint with the at least one related concept.

112. The method of claim 111, wherein the query is a document.

113. The method of claim 111, wherein the fingerprint is a list of concepts and their associated relevance weights.

114. The method of claim 111, wherein the terminology system is a thesaurus.

115. The method of claim 111, wherein expanding the query comprises adding the at least one related concept to the query.

116. The method of claim 111, further comprising querying a knowledge base with the expanded query.

117. A system for knowledge discovery comprising:

a memory configured for storing textual data;

a processor, coupled to the memory, configured for performing steps comprising,

receiving a query;

generating a fingerprint for the query, wherein the fingerprint comprises at least one concept;

determining at least one related concept to the fingerprint according to a terminology system; and

expanding the fingerprint with the at least one related concept.

118. The system of claim 117, wherein the query is a document.

119. The system of claim 117, wherein the fingerprint is a list of concepts and their associated relevance weights.

120. The system of claim 117, wherein the terminology system is a thesaurus.

121. The system of claim 117, wherein expanding the query comprises adding the at least one related concept to the query.

122. The system of claim 117, further comprising querying a knowledge base with the expanded query.

123. A computer readable medium with computer executable instructions embodied thereon for knowledge discovery comprising:

receiving a query;

generating a fingerprint for the query, wherein the fingerprint comprises at least one concept;

determining at least one related concept to the fingerprint according to a terminology system; and

expanding the fingerprint with the at least one related concept.

124. The computer readable medium of claim 123, wherein the query is a document.

125. The computer readable medium of claim 123, wherein the fingerprint is a list of concepts and their associated relevance weights.

126. The computer readable medium of claim 123, wherein the terminology system is a thesaurus.

127. The computer readable medium of claim 123, wherein expanding the query comprises adding the at least one related concept to the query.

128. The computer readable medium of claim 123, further comprising querying a knowledge base with the expanded query.

129. A method for homograph disambiguation comprising:

receiving a fingerprint;

determining if the fingerprint comprises one or more concepts that are homographs; and

for each homograph, determining a matching score between the fingerprint and each of a set of context fingerprints associated with a different meaning of the homograph wherein a meaning of the homograph is selected from the set of context fingerprints by choosing the context fingerprint with a maximum matching score if that score is above a threshold.

130. The method of claim 129, wherein the fingerprint is a list of concepts and their associated relevance weights.

131. The method of claim 129, wherein a homograph is one of a group of words that share the same spelling but have different meanings.

132. The method of claim 129, wherein a context fingerprint is a list of concepts and their associated relevance weights which is constructed based on co-occurrence of concepts in documents with the concept the context fingerprint is created for.

133. The method of claim 129, wherein determining a matching score comprises performing a matching algorithm.

134. The method of claim 133, wherein performing a matching algorithm comprises:

storing each context fingerprint as a vector; and

performing a vector matching algorithm.

135. A system for homograph disambiguation comprising:

a memory configured for storing textual data;

a processor, coupled to the memory, configured for performing steps comprising,

receiving a fingerprint;

determining if the fingerprint comprises one or more concepts that are homographs; and

for each homograph, determining a matching score between the fingerprint and each of a set of context fingerprints associated with a different meaning of the homograph wherein a meaning of the homograph is selected from the set of context fingerprints by choosing the context fingerprint with a maximum matching score if that score is above a threshold.

136. The system of claim 135, wherein the fingerprint is a list of concepts and their associated relevance weights.

137. The system of claim 135, wherein a homograph is one of a group of words that share the same spelling but have different meanings.

138. The system of claim 135, wherein a context fingerprint is a list of concepts and their associated relevance weights which is constructed based on co-occurrence of concepts in documents with the concept the context fingerprint is created for.

139. The system of claim 135, wherein determining a matching score comprises performing a matching algorithm.

140. The system of claim 139, wherein performing a matching algorithm comprises:

storing each context fingerprint as a vector; and

performing a vector matching algorithm.

141. A computer readable medium with computer executable instructions embodied thereon for homograph disambiguation comprising:

receiving a fingerprint;

determining if the fingerprint comprises one or more concepts that are homographs; and

for each homograph, determining a matching score between the fingerprint and each of a set of context fingerprints associated with a different meaning of the homograph wherein a meaning of the homograph is selected from the set of context fingerprints by choosing the context fingerprint with a maximum matching score if that score is above a threshold.

142. The computer readable medium of claim 141, wherein the fingerprint is a list of concepts and their associated relevance weights.

143. The computer readable medium of claim 141, wherein a homograph is one of a group of words that share the same spelling but have different meanings.

144. The computer readable medium of claim 141, wherein a context fingerprint is a list of concepts and their associated relevance weights which is constructed based on co-occurrence of concepts in documents with the concept the context fingerprint is created for.

145. The computer readable medium of claim 141, wherein determining a matching score comprises performing a matching algorithm.

146. The computer readable medium of claim 145, wherein performing a matching algorithm comprises:

storing each context fingerprint as a vector; and

performing a vector matching algorithm.

147. A method for knowledge discovery comprising:

receiving a result set for a query;

determining one or more concepts in the result set that are not part of the query; and

presenting the one or more concepts to a user in a structured overview for inclusion in a subsequent query.

148. The method of claim 147, further comprising performing a search with the query.

149. The method of claim 147, wherein presenting the one or more concepts to the user comprises displaying the one or more concepts on a display device.

150. The method of claim 147, wherein the structured overview is a grouping of concepts according to a terminology system (TS).

151. The method of claim 150, wherein the grouping utilizes a hierarchy defined in the TS.

152. The method of claim 150, wherein the grouping utilizes semantic types defined in the TS.

153. The method of claim 147, wherein presenting the one or more concepts to the user comprises permitting the user to select any of the one or more concepts to add to the subsequent query.

154. The method of claim 147, further comprising performing the subsequent query.

155. The method of claim 154, further comprising repeating the steps of claim 147 after performing the subsequent query.

156. A system for knowledge discovery comprising:

a memory configured for storing textual data;

a processor, coupled to the memory, configured for performing steps comprising,

receiving a result set for a query;

determining one or more concepts in the result set that are not part of the query; and

presenting the one or more concepts to a user in a structured overview for inclusion in a subsequent query.

157. The system of claim 156, further comprising performing a search with the query.

158. The system of claim 156, wherein presenting the one or more concepts to the user comprises displaying the one or more concepts on a display device.

159. The system of claim 156, wherein the structured overview is a grouping of concepts according to a terminology system (TS).

160. The system of claim 159, wherein the grouping utilizes a hierarchy defined in the TS.

161. The system of claim 159, wherein the grouping utilizes semantic types defined in the TS.

162. The system of claim 156, wherein presenting the one or more concepts to the user comprises permitting the user to select any of the one or more concepts to add to the subsequent query.

163. The system of claim 156, further comprising performing the subsequent query.

164. The system of claim 163, further comprising repeating the steps of claim 156 after performing the subsequent query.

165. A computer readable medium with computer executable instructions embodied thereon for knowledge discovery comprising:

receiving a result set for a query;

determining one or more concepts in the result set that are not part of the query; and

presenting the one or more concepts to a user in a structured overview for inclusion in a subsequent query.

166. The computer readable medium of claim 165, further comprising performing a search with the query.

167. The computer readable medium of claim 165, wherein presenting the one or more concepts to the user comprises displaying the one or more concepts on a display device.

168. The computer readable medium of claim 165, wherein the structured overview is a grouping of concepts according to a terminology system (TS).

169. The computer readable medium of claim 168, wherein the grouping utilizes a hierarchy defined in the TS.

170. The computer readable medium of claim 168, wherein the grouping utilizes semantic types defined in the TS.

171. The computer readable medium of claim 165, wherein presenting the one or more concepts to the user comprises permitting the user to select any of the one or more concepts to add to the subsequent query.

172. The computer readable medium of claim 165, further comprising performing the subsequent query.

173. The computer readable medium of claim 172, further comprising repeating the steps of claim 165 after performing the subsequent query.

174. A method for knowledge discovery comprising:

searching a body of knowledge by metadata and by vector matching; and

displaying a result set of the searching.

175. The method of claim 174, wherein the body of knowledge is a collection of fingerprints.

176. The method of claim 174, wherein searching the body of knowledge by metadata comprises performing a Boolean search.

177. The method of claim 174, wherein searching the body of knowledge by metadata comprises performing a search by determining a deviation of a metadata value from a specified value and expressing the deviation in a relevance score.

178. The method of claim 174, wherein searching the body of knowledge by vector matching comprises:

storing each context fingerprint as a vector; and

performing a vector matching algorithm.

179. The method of claim 174, wherein searching the body of knowledge by metadata and by vector matching is performed simultaneously.

180. The method of claim 174, wherein searching the body of knowledge by metadata and by vector matching is performed sequentially.

181. The method of claim 174, wherein displaying a result set of the searching comprises displaying the result set on a display device.

182. A system for knowledge discovery comprising:

a memory configured for storing textual data;

a processor, coupled to the memory, configured for performing steps comprising,

searching a body of knowledge by metadata and by vector matching; and

displaying a result set of the searching.

183. The system of claim 182, wherein the body of knowledge is a collection of fingerprints.

184. The system of claim 182, wherein searching the body of knowledge by metadata comprises performing a Boolean search.

185. The system of claim 182, wherein searching the body of knowledge by metadata comprises performing a search by determining a deviation of a metadata value from a specified value and expressing the deviation in a relevance score.

186. The system of claim 182, wherein searching the body of knowledge by vector matching comprises:

storing each context fingerprint as a vector; and

performing a vector matching algorithm.

187. The system of claim 182, wherein searching the body of knowledge by metadata and by vector matching is performed simultaneously.

188. The system of claim 182, wherein searching the body of knowledge by metadata and by vector matching is performed sequentially.

189. The system of claim 182, wherein displaying a result set of the searching comprises displaying the result set on a display device.

190. A computer readable medium with computer executable instructions embodied thereon for knowledge discovery comprising:

searching a body of knowledge by metadata and by vector matching; and

displaying a result set of the searching.

191. The computer readable medium of claim 190, wherein the body of knowledge is a collection of fingerprints.

192. The computer readable medium of claim 190, wherein searching the body of knowledge by metadata comprises performing a Boolean search.

193. The computer readable medium of claim 190, wherein searching the body of knowledge by metadata comprises performing a search by determining a deviation of a metadata value from a specified value and expressing the deviation in a relevance score.

194. The computer readable medium of claim 190, wherein searching the body of knowledge by vector matching comprises:

storing each context fingerprint as a vector; and

performing a vector matching algorithm.

195. The computer readable medium of claim 190, wherein searching the body of knowledge by metadata and by vector matching is performed simultaneously.

196. The computer readable medium of claim 190, wherein searching the body of knowledge by metadata and by vector matching is performed sequentially.

197. The computer readable medium of claim 190, wherein displaying a result set of the searching comprises displaying the result set on a display device.

198. A method for knowledge discovery comprising:

plotting a set of concepts out of a selected set of fingerprints on a terminology system wherein a map is generated;

selecting a concept out of the set of concepts;

displaying the map to the user; and

indicating, to the user, a relative importance of the selected concept to the set of concepts.

199. The method of claim 198, wherein each fingerprint represents a document.

200. The method of claim 198, wherein each fingerprint represents a person.

201. The method of claim 198, wherein each fingerprint represents an organization.

202. The method of claim 198, wherein indicating, to the user, the relative importance of the selected concept to the set of concepts comprises displaying the selected concept in a different color than that of the other concepts.

203. The method of claim 198, wherein indicating, to the user, the relative importance of the selected concept to the set of concepts comprises displaying the selected concept with an object larger than that of the other concepts.

204. A system for knowledge discovery comprising:

a memory configured for storing textual data;

a processor, coupled to the memory, configured for performing steps comprising,

plotting a set of concepts out of a selected set of fingerprints on a terminology system wherein a map is generated;

selecting a concept out of the set of concepts;

displaying the map to the user; and

indicating, to the user, a relative importance of the selected concept to the set of concepts.

205. The system of claim 204, wherein each fingerprint represents a document.

206. The system of claim 204, wherein each fingerprint represents a person.

207. The system of claim 204, wherein each fingerprint represents an organization.

208. The system of claim 204, wherein indicating, to the user, the relative importance of the selected concept to the set of concepts comprises displaying the selected concept in a different color than that of the other concepts.

209. The system of claim 204, wherein indicating, to the user, the relative importance of the selected concept to the set of concepts comprises displaying the selected concept with an object larger than that of the other concepts.

210. A computer readable medium with computer executable instructions embodied thereon for knowledge discovery comprising:

plotting a set of concepts out of a selected set of fingerprints on a terminology system wherein a map is generated;

selecting a concept out of the set of concepts;

displaying the map to the user; and

indicating, to the user, a relative importance of the selected concept to the set of concepts.

211. The computer readable medium of claim 210, wherein each fingerprint represents a document.

212. The computer readable medium of claim 210, wherein each fingerprint represents a person.

213. The computer readable medium of claim 210, wherein each fingerprint represents an organization.

214. The computer readable medium of claim 210, wherein indicating, to the user, the relative importance of the selected concept to the set of concepts comprises displaying the selected concept in a different color than that of the other concepts.

215. The computer readable medium readable medium of claim 210, wherein indicating, to the user, the relative importance of the selected concept to the set of concepts comprises displaying the selected concept with an object larger than that of the other concepts.

216. A method for knowledge discovery comprising:

receiving a plurality of fingerprints for a document, wherein each of the plurality of fingerprints is generated from a unique source;

receiving, from a user, an indication of relative importance of the plurality of fingerprints;

searching a body of knowledge with the plurality of fingerprints based on the relative importance; and

displaying a result set to the user.

217. The method of claim 216, wherein each of the plurality of fingerprints is a list of concepts and their associated relevance weight.

218. The method of claim 216, wherein the unique source comprises a freetext thesaurus.

219. The method of claim 216, wherein the unique source comprises a terminology system.

220. The method of claim 216, wherein the unique source comprises a combination of unique sources.

221. The method of claim 220, wherein one of the plurality of fingerprints was generated from a first terminology system and another one of the plurality of fingerprints was generated from a second terminology system.

222. The method of claim 221, wherein the first terminology system and the second terminology system are from different technical fields.

223. The method of claim 216, wherein the plurality of fingerprints were each generated from a different part of the document.

224. The method of claim 216, wherein the body of knowledge is a collection of fingerprints.

225. A system for knowledge discovery comprising:

a memory configured for storing textual data;

a processor, coupled to the memory, configured for performing steps comprising,

receiving a plurality of fingerprints for a document, wherein each of the plurality of fingerprints is generated from a unique source;

receiving, from a user, an indication of relative importance of the plurality of fingerprints;

searching a body of knowledge with the plurality of fingerprints based on the relative importance; and

displaying a result set to the user.

226. The system of claim 225, wherein each of the plurality of fingerprints is a list of concepts and their associated relevance weight.

227. The system of claim 225, wherein the unique source comprises a freetext thesaurus.

228. The system of claim 225, wherein the unique source comprises a terminology system.

229. The system of claim 225, wherein the unique source comprises a combination of unique sources.

230. The system of claim 229, wherein one of the plurality of fingerprints was generated from a first terminology system and another one of the plurality of fingerprints was generated from a second terminology system.

231. The system of claim 230, wherein the first terminology system and the second terminology system are from different technical fields.

232. The system of claim 225, wherein the plurality of fingerprints were each generated from a different part of the document.

233. The system of claim 225, wherein the body of knowledge is a collection of fingerprints.

234. A computer readable medium with computer executable instructions embodied thereon for knowledge discovery comprising:

receiving a plurality of fingerprints for a document, wherein each of the plurality of fingerprints is generated from a unique source;

receiving, from a user, an indication of relative importance of the plurality of fingerprints;

searching a body of knowledge with the plurality of fingerprints based on the relative importance; and

displaying a result set to the user.

235. The computer readable medium of claim 234, wherein each of the plurality of fingerprints is a list of concepts and their associated relevance weight.

236. The computer readable medium of claim 234, wherein the unique source comprises a freetext thesaurus.

237. The computer readable medium of claim 234, wherein the unique source comprises a terminology system.

238. The computer readable medium of claim 234, wherein the unique source comprises a combination of unique sources.

239. The computer readable medium of claim 238, wherein one of the plurality of fingerprints was generated from a first terminology system and another one of the plurality of fingerprints was generated from a second terminology system.

240. The computer readable medium of claim 239, wherein the first terminology system and the second terminology system are from different technical fields.

241. The computer readable medium of claim 234, wherein the plurality of fingerprints were each generated from a different part of the document.

242. The computer readable medium of claim 234, wherein the body of knowledge is a collection of fingerprints.