# US 11,443,101 B2 · Flexible pseudo-parsing of dense semi-structured text > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,443,101 B2 - Document ID: US11443101B2 - Status: Granted - Kind: B2 - Application number: 17/087,994 - Filed: 2020-11-03 - Published: 2022-05-05 - Granted: 2022-09-13 - Earliest claimed: 2020-11-03 - Assignees: International Business Machine Corporation - Applicants: International Business Machines Corporation - Inventors: Christopher F. Ackermann; Charles E. Beller; Michael Drzewucki - Web page: https://patentgrep.com/patent/US11443101B2 ## CPC - G06F 40/169: Annotation, e.g. comment data or footnotes - G06F 40/279: Recognition of textual entities - G06F 16/81: Indexing, e.g. XML tags; Data structures therefor; Storage structures - G06F 16/84: Mapping; Conversion - G06F 40/186: Templates - G06F 40/205: Parsing - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract An embodiment for extracting information from semi-structured text is provided. The embodiment may include identifying one or more high confidence alignments of one or more entities and identifiers in a set of documents. The embodiment may also include analyzing one or more blocks of semi-structured text containing the one or more entities and identifiers. The embodiment may further include identifying one or more known alignments in each of the one or more blocks of semi-structured text. The embodiment may also include generating a structure template. The embodiment may further include applying the structure template to each of the one or more blocks of semi-structured text. The embodiment may also include annotating the set of documents with metadata reflecting the structure template and a location of each of the one or more blocks of semi-structured text. ## Claims ### Claim 1 A computer-based method of extracting information from semi-structured text, the method comprising: identifying one or more high confidence alignments of one or more entities and one or more identifiers in a set of documents, wherein the one or more high confidence alignments are a string of text within a maximum pre-configured proximity threshold, wherein the maximum pre-configured proximity threshold of the one or more high confidence alignments is a threshold where only one entity and one identifier occur in each of the one or more blocks of semi-structured text, and the one entity and the one identifier are within twenty characters of each other; analyzing one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers; identifying one or more known alignments in each of the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers based on the one or more high confidence alignments, wherein the one or more known alignments are a string of text; generating a structure template based on the one or more known alignments; applying the structure template to each of the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers; and annotating the set of documents with metadata, wherein the metadata annotation in the set of documents includes the structure template placed adjacent to each of the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers, wherein the metadata annotation in the set of documents includes a location of each of the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers, and wherein the metadata annotation in the set of documents is stored in a permanent query-specific knowledge base. ### Claim 2 (depends on 1) The method of claim 1, wherein analyzing the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers further comprises: identifying the one or more blocks of semi-structured text where a plurality of entities and a plurality of identifiers are mixed together. ### Claim 3 (depends on 2) The method of claim 2, wherein the plurality of entities and the plurality of identifiers mixed together are contained in a semi-structured block of text where at least 80% of the text is composed of entities and identifiers associated with the entities. ### Claim 4 (depends on 3) The method of claim 3, further comprising: populating a temporary query-specific knowledge base based on the one or more high confidence alignments. ### Claim 5 (depends on 4) The method of claim 4, wherein the temporary query-specific knowledge base is added to a permanent structured knowledge base. ### Claim 6 A computer system, the computer system comprising: one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage medium, and program instructions stored on at least one of the one or more tangible storage medium for execution by at least one of the one or more processors via at least one of the one or more memories, wherein the computer system is capable of performing a method comprising: identifying one or more high confidence alignments of one or more entities and one or more identifiers in a set of documents, wherein the one or more high confidence alignments are a string of text within a maximum pre-configured proximity threshold, wherein the maximum pre-configured proximity threshold of the one or more high confidence alignments is a threshold where only one entity and one identifier occur in each of the one or more blocks of semi-structured text, and the one entity and the one identifier are within twenty characters of each other; analyzing one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers; identifying one or more known alignments in each of the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers based on the one or more high confidence alignments, wherein the one or more known alignments area string of text; generating a structure template based on the one or more known alignments; applying the structure template to each of the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers; and annotating the set of documents with metadata, wherein the metadata annotation in the set of documents includes the structure template placed adjacent to each of the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers, wherein the metadata annotation in the set of documents includes a location of each of the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers, and wherein the metadata annotation in the set of documents is stored in a permanent query-specific knowledge base. ### Claim 7 (depends on 6) The computer system of claim 6, wherein analyzing the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers further comprises: identifying the one or more blocks of semi-structured text where a plurality of entities and a plurality of identifiers are mixed together. ### Claim 8 (depends on 7) The computer system of claim 7, wherein the plurality of entities and the plurality of identifiers mixed together are contained in a semi-structured block of text where at least 80% of the text is composed of entities and identifiers associated with the entities. ### Claim 9 (depends on 8) The computer system of claim 8, further comprising: populating a temporary query-specific knowledge base based on the one or more high confidence alignments. ### Claim 10 (depends on 9) The computer system of claim 9, wherein the temporary query-specific knowledge base is added to a permanent structured knowledge base. ### Claim 11 A computer program product, the computer program product comprising: one or more computer-readable tangible storage medium and program instructions stored on at least one of the one or more tangible storage medium, the program instructions executable by a processor capable of performing a method, the method comprising: identifying one or more high confidence alignments of one or more entities and one or more identifiers in a set of documents, wherein the one or more high confidence alignments are a string of text within a maximum pre-configured proximity threshold, wherein the maximum pre-configured proximity threshold of the one or more high confidence alignments is a threshold where only one entity and one identifier occur in each of the one or more blocks of semi-structured text, and the one entity and the one identifier are within twenty characters of each other; analyzing one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers; identifying one or more known alignments in each of the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers based on the one or more high confidence alignments, wherein the one or more known alignments are a string of text; generating a structure template based on the one or more known alignments; applying the structure template to each of the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers; and annotating the set of documents with metadata, wherein the metadata annotation in the set of documents includes the structure template placed adjacent to each of the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers, wherein the metadata annotation in the set of documents includes a location of each of the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers, and wherein the metadata annotation in the set of documents is stored in a permanent query-specific knowledge base. ### Claim 12 (depends on 11) The computer program product of claim 11, wherein analyzing the one or more blocks of semi-structured text containing the one or more entities and the one or more identifiers further comprises: identifying the one or more blocks of semi-structured text where a plurality of entities and a plurality of identifiers are mixed together. ### Claim 13 (depends on 12) The computer program product of claim 12, wherein the plurality of entities and the plurality of identifiers mixed together are contained in a semi-structured block of text where at least 80% of the text is composed of entities and identifiers associated with the entities. ### Claim 14 (depends on 13) The computer program product of claim 13, further comprising: populating a temporary query-specific knowledge base based on the one or more high confidence alignments.