WO2015199581A2 - Procédé de transformation préliminaire d'une collection initiale de données, procédé de formation d'une carte de liens des composants de parties de structures logiques d'une collection de données initiale structurée qui a été transformée, procédé de recherche dans une collection de données initiale structurée qui a été transformée utilisant la carte de liens des composants et systèmes et dispositfs pour la mise en oeuvre de ces procédés - Google Patents

Procédé de transformation préliminaire d'une collection initiale de données, procédé de formation d'une carte de liens des composants de parties de structures logiques d'une collection de données initiale structurée qui a été transformée, procédé de recherche dans une collection de données initiale structurée qui a été transformée utilisant la carte de liens des composants et systèmes et dispositfs pour la mise en oeuvre de ces procédés Download PDF

Info

Publication number
WO2015199581A2
WO2015199581A2 PCT/RU2015/000392 RU2015000392W WO2015199581A2 WO 2015199581 A2 WO2015199581 A2 WO 2015199581A2 RU 2015000392 W RU2015000392 W RU 2015000392W WO 2015199581 A2 WO2015199581 A2 WO 2015199581A2
Authority
WO
WIPO (PCT)
Prior art keywords
logical
data structure
elements
grammatically
data
Prior art date
Application number
PCT/RU2015/000392
Other languages
English (en)
Russian (ru)
Other versions
WO2015199581A3 (fr
Inventor
Игорь Петрович РОГАЧЕВ
Original Assignee
Игорь Петрович РОГАЧЕВ
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from RU2014126198/08A external-priority patent/RU2571406C1/ru
Priority claimed from RU2014126195/08A external-priority patent/RU2572367C1/ru
Priority claimed from RU2014126199/08A external-priority patent/RU2571407C1/ru
Priority claimed from RU2014126197/08A external-priority patent/RU2571405C1/ru
Application filed by Игорь Петрович РОГАЧЕВ filed Critical Игорь Петрович РОГАЧЕВ
Priority to EA201700031A priority Critical patent/EA201700031A1/ru
Publication of WO2015199581A2 publication Critical patent/WO2015199581A2/fr
Publication of WO2015199581A3 publication Critical patent/WO2015199581A3/fr

Links

Definitions

  • the group of inventions relates to solutions in the field of processing data arrays, in particular, to solutions in the field of processing structured data arrays containing natural language text, and can be used for the necessary preliminary transformation of a structured data array, forming a map of connections of components of logical structure parts of a transformed structured data array and search for relevant information in a structured data array using information from a communication map of parts of logical constructions of a structured data array.
  • a known method of displaying a document described in US patent 8335754 B2, G06F17 / 00, 12/18/2012 (TAGGED INC ET AL).
  • a known method is to access a data array containing text data; extracting at least one sentence from said array; constructing a tree of relations identifying parts of speech and grammatical relations in the structure of the tree; the formation of semantic structures, each of which contains three elements from the aforementioned tree structure - subject-predicate-object; preserving the semantic network of said tree structure with connections; and displaying relevant data, which can be made in the form of: a concept map, a list of facts, text fragments, labels, tags or annotations.
  • the disadvantage of this method is that when forming the aforementioned tree structure of relations, the subject-predicate-object relationships do not generate logical structures consisting of logical sections containing grammatically and spelling-correct elements, which makes it impossible to construct a correct map of the connections of components and terms of logical structures.
  • the known method is carried out by creating a preliminary data directory with a reduced number of spelling and typographical errors in the text of the data directory, and is provided due to the fact that they use descriptive parameters of the text terms in the form of a prefix string, and the above-mentioned catalog contains prefixes and or fragments of terms words along with data containing spelling errors, which allows for the analysis of the specified data array by comparing with the terms of the user's query to present relevant antenna data regardless of their correct presentation in the text structure.
  • the disadvantage of this method is that when forming the aforementioned preliminary data catalog they do not provide the formation of logical structures containing grammatically and spelling correct semantic parts of logical sections and do not provide a map of connections of components of the mentioned grammatically and spelling correct semantic parts of logical sections of logical sections of logical structures of logical structures, which does not provide the required search accuracy in the converted data array.
  • the closest analogue (prototype) of the claimed solution is a method for automated processing of natural language text by means of its semantic indexing, described in patent RU 2399959 C2, G09B19 / 00, September 20, 2010 (CLOSED JOINT STOCK COMPANY "AVIKOMP SERVISES”).
  • the known method is a method in which text is segmented in electronic form into elementary units, identifies stable phrases, form sentences, identifies semantically significant objects and semantically significant relationships between them, form many triads for each semantically significant relationship, in which a single triad of the first type corresponds to a relationship established by a semantically significant relationship between two semantically significant objects, each of the triads of the second type of co corresponds to the value of a specific attribute of one of these semantically significant objects, each of the triads of the third type corresponds to the value of a specific attribute of the semantically significant relation, index all related semantically meaningful relations separately from the set of triads, store the generated triads and obtained indices in the database together with a link to the source text from which these triads are formed.
  • the task to which the claimed invention is directed is to provide such preliminary processing of a structured data array containing natural language text that would allow further transformations to be carried out and generate a logical, grammatical and spelling correct resulting structure containing logical constructions array elements and provides quick and convenient navigation through array elements.
  • the technical result is the preliminary conversion of a structured data array suitable for the further formation of a logical, grammatically and spelling-correct data structure.
  • the claimed technical result is achieved due to the fact that they perform a method of converting a structured source data array containing at least natural language text, said method comprising at least the steps of:
  • structurally complex linguistic constructions which are structurally complex linguistic constructions containing contextual terms and structurally complex linguistic constructions that do not contain contextual terms
  • structurally simple language constructs which are structurally simple language constructs containing contextual terms, and structurally simple language constructs that do not contain contextual terms
  • G form the final data structure of the structured initial data array containing the elements of the final data structure, and the elements of the final data structure are structurally simple language constructs that do not contain contextual terms and obtained grammatically and spelling structurally correct due to transformations of the first, second and third types -simple language constructs that do not contain contextual terms.
  • Embodiments of the present invention relate to methods, devices, systems, and computer-readable media using the method described above.
  • Figure 1 shows a General diagram of the steps of the claimed method of converting a structured source data array containing at least natural language text in accordance with the first embodiment the implementation of the present invention.
  • FIG. 2 depicts a general flowchart for the steps of the claimed method for converting a structured source data array containing at least natural language text in accordance with a third embodiment of the present invention.
  • FIG. 3 shows a general diagram of the step of generating the first data structure.
  • FIG. 4 shows the general structure of the original data structure from which the first data structure is formed.
  • Fig 5 shows a General diagram of the stage of forming a database of logical connections of logical partitions.
  • FIG. 6 shows the general principle of forming a database of logical connections of logical partitions.
  • Fig. 7 shows a general diagram of a step for generating a second data structure.
  • FIG. 8 shows the general structure of a second data structure.
  • FIG. 9 shows the general scheme of the stage of formation of the database of semantic parts.
  • FIG. 10 depicts the general principle of forming a database of semantic parts.
  • FIG. 11 shows a general diagram of the stage of formation of grammatically and spelling-correct semantic parts
  • FIG. 12 shows a general diagram of a second data structure obtained after the step of forming grammatically and spelling-correct semantic parts.
  • FIG. 13 shows a general diagram of the step of generating the resulting data structure.
  • FIG. 14 shows a general diagram of the resulting data structure.
  • FIG. 15 shows the general structure of the elements of the resulting data structure.
  • FIG. 16 shows a General diagram of the stages of the claimed method of forming a map of the relations of the components of grammatically and spelling correct semantic parts of logical sections of logical structures of the transformed structured source data array.
  • FIG. 17-18 depict a general diagram of the steps of the claimed method for searching for relevant information in a transformed structured source data array containing at least logical constructions containing grammatically and spelling-correct semantic parts of logical sections of logical constructions.
  • FIG. 19 shows a general diagram of a system containing a device for implementing the claimed methods for converting a structured source data array, methods for generating a map of the components of a transformed structured source data array, and methods for finding relevant information in a converted structured source data array.
  • the claimed method will be considered on the example of processing a structured data array containing natural language text, which is, without limitation, regulatory legal acts (NPA). It should be obvious to a person skilled in the art that despite the fact that in this particular implementation example of the present invention, the conversion of NLA is carried out, such a conversion method can be applied to any structured data array similar to NLA.
  • NPA regulatory legal acts
  • NLAs are of a law-making nature: in them, the rules of law are either established, changed or canceled.
  • Normative legal acts are carriers of legal norms;
  • the regulatory legal act contains legal instruments with the help of which the legal regulatory influence is implemented.
  • NPA is published only within the competence of the law-making body; 4) NPA is clothed in documentary form and has the following details: type of regulatory act, its name, authority that adopted it, date, place of adoption of the act, number;
  • NPA is not a chaotic set of provisions (proposals), but has a certain structure
  • the NLA must be consistent with the constitution or another higher NLA, which has greater legal force.
  • NPA must be brought to the attention of citizens and organizations, i.e. publication, and only after that the state has the right to demand its rigorous execution, based on the presumption of knowledge of the law, and impose sanctions for its failure.
  • structured data array within the framework of the claimed invention can be considered not only a set of legal acts, but also a separate independent legal acts, which is, for example: the Constitution, law, decree, regulation, etc.
  • a separate LA can, for example, consist of parts, chapters, sections, articles.
  • the instrument of the legal regulatory impact of the legal acts is the legal rule, denounced in the structure of the regulatory requirement, which in turn is an element (part) of the rule of law (legal norm).
  • a structured source data array containing at least natural language text is, in particular, a legal entity, which is a text consisting of structurally simple and structurally complex language constructs.
  • a structurally simple linguistic construction is understood to mean complete text fragments — grammatically organized word combinations, i.e. offers. Moreover, each sentence is necessarily characterized by semantic completeness.
  • structurally complex linguistic construction is understood as constructions of the “listing category” type.
  • the main goal of the preliminary transformation of a structured source data array is the formation of a structured data array consisting only of sentences. This goal is realized by converting structurally complex linguistic constructions into structurally simple linguistic constructions - sentences.
  • Another purpose of the preliminary conversion is to clarify contextual terms and simplify sentences containing contextual terms.
  • contextual terms are understood to mean the conditional naming of such terms that either do not have an independent semantic meaning outside the context, or have many semantic meanings in the context.
  • the context refers to a relatively finished meaningful passage of text within which the meaning and meaning of a single word, phrase or combination of phrases, as well as a group of at least two sentences, in two different passages are most accurately and clearly identified. text, provided that in one of the sentences there is a link to another sentence.
  • the present invention provides a method for converting a structured source data array containing at least natural language text, said method comprising at least the steps of:
  • structurally complex linguistic constructions which are structurally complex linguistic constructions containing contextual terms and structurally complex linguistic constructions that do not contain contextual terms;
  • structurally simple language constructs which are structurally simple language constructs containing contextual terms, and structurally simple language constructs that do not contain contextual terms;
  • G form the final data structure of the structured source data array containing the elements of the final data structure, and the elements of the final data structures are structurally simple language constructs that do not contain unite terms and obtained by transformations of the first, second and third types of grammatically and spelling correct structurally simple language constructs that do not contain contextual terms.
  • -conversion of the first type includes format conversion and linguistic conversion
  • -Transformation of the third type includes format conversion, contextual conversion and linguistic conversion.
  • a third possible embodiment of the present invention there is provided a method according to any one of the first or second embodiments, characterized in that it further comprises at least the steps of:
  • the first data structure of the structured source data array is formed containing the elements of the first data structure, said elements of the first data structure containing the first logical partitions and second logical partitions;
  • K form a second data structure of a structured source data array containing the elements of said second data structure, said elements of the second data structure containing logical structures of logical partitions of said elements of the first data structure formed using information from said logical linking database of logical partitions, said logical sections contain the first semantic parts and the second semantic parts;
  • K form a database of semantic parts of logical sections from said second semantic parts, wherein said second semantic parts are excluded from the corresponding mentioned logical sections;
  • step 3 is characterized by at least steps in which:
  • step I) is characterized by at least steps in which:
  • the elements of the first data structure that do not have a logical connection between logical partitions are identified; and - form a database of logical relationships of logical sections of the elements of the first data structure.
  • step K) is characterized by at least the stages in which.
  • step L) is characterized by at least the stages in which.
  • At least special semantic parts of the first logical sections of the elements of the second data structure and the special semantic parts of the second logical sections of the elements of the second data structure are identified and form a database of special semantic parts of the logical sections of the elements of the second data structures by moving said special semantic parts to said logically generated database of special semantic parts Partition elements of the second data structure.
  • step M) is characterized by at least steps in which:
  • the mentioned second logical sections of the elements of the second data structure identify at least the refinement structures of the second semantic parts of the second logical sections;
  • step H is characterized by at least steps in which:
  • a method according to a ninth embodiment characterized in that said logical constructions from said resultant data structure may further comprise said semantic combinations of grammatically and spelling-correct semantic parts of second logical sections of elements of the second data structure.
  • I) identify elements of the first data structure containing one mentioned first logical partition, and elements of the first data structure containing one mentioned second logical partition; identifying elements of a first data structure containing more than one of said first logical partitions, and elements of a first data structure containing more than one of said second logical partition; in the elements of the first data structure containing more than one of said first logical partitions and in the elements of the first data structure containing more than one of said second logical partitions, logical relationships between said first logical partitions or between said second logical partitions are identified; in the elements of the first data structure containing more than one of the first logical partitions and in the elements of the first data structure containing more than one of the second logical partitions, elements of the first data structure that do not have a logical connection between logical partitions are identified; and form a database of logical relationships of logical partitions of the elements of the first data structure;
  • K form logical structures of logical partitions of elements of the first data structure using information from a database of logical connections of logical partitions of elements of the first data structure and logical partitions of said elements of the first data structure containing one of the first logical partitions and logical partitions of said elements of the first data structure, containing one said second logical partition; and form a second data structure of the original structured data array containing elements of the second data structure, said elements of the second data structure being formed logical structures of logical partitions of the first data structure;
  • K identify the first logical partitions of the elements of the second data structure and the second logical partitions of the elements of the second data structure; in said first logical sections and second logical sections of the elements of the second data structure, the first semantic parts and the second semantic parts are identified; and in said first and second logical sections of the elements of the second data structure, at least the special semantic parts of the first logical sections of the elements of the second data structure and the special semantic parts of the second logical sections of the elements of the second data structure are identified and form the database special semantic parts of the logical sections of the elements of the second data structure by moving said special semantic parts to the said generated database of special semantic parts of the logical sections of the elements of the second data structure;
  • H form semantic combinations of grammatically and spelling-correct semantic parts of the second logical sections of the elements of the third structure from the first grammatically and spelling-correct semantic parts of the second logical sections of the elements of the second data structure and the grammatically and spelling-correcting refining structures of the second semantic parts of the second logical sections of the elements of the second data structure data; and form the resulting data structure of the original structured data array containing the elements of the resulting data structure, said elements of the resulting data structure being logical constructs containing the grammatically and spelling-correct semantic parts of the logical sections of the elements of the second data structure.
  • a method according to the eleventh embodiment characterized in that said logical constructs from said resultant data structure may further comprise said semantic combinations of grammatically and spelling-correct semantic parts of second logical sections of elements of the second data structure.
  • a device for converting a structured source data array comprising at least:
  • processors one or more processors
  • a device in a fourteenth possible embodiment of the present invention, characterized in that said one or more structured source data arrays to be converted are downloadable, and said device is capable of connecting to a database in which said downloadable one to be converted one is stored or several structured source data arrays for loading into said memory devices of at least one loadable, structured source data array to be converted.
  • a system for converting a structured source data array comprising at least:
  • one or more servers providing regulation of data exchange in said system
  • one or more databases for storing data configured to interact with said one or more devices
  • a computer-readable storage medium comprising a program code that, upon execution, causes a processor or processors of a device with which the computer-readable storage medium interacts to perform the steps of a method as in any of the first to twelfth embodiments.
  • a method of forming a map of links of components of grammatically and spelling correct semantic parts of logical sections of logical constructions transformed by a method according to any one of the first to twelfth structured source data arrays the method forming a map of communications includes at least the stages in which:
  • each grammatically and spelling-correct semantic part containing said components containing at least one said component, said component containing no more than one component value, and form a component table containing at least the mentioned identifiers ized components and their meanings,
  • the concepts are identified that semantically coincide with the mentioned component values contained in the component combination map, and a table of semantically matching concepts containing at least the component values is formed contained in the aforementioned map of combinations of components, and semantically coinciding concepts with them;
  • the present invention provides a method for searching for relevant information in a transformed by a method according to any one of the first to twelfth structured source data array, the method for searching for relevant information comprising at least the steps of:
  • search query containing at least one search term
  • the present invention provides a method for searching for relevant information in a transformed by a method according to any one of the first to twelfth structured source data array, the method for searching for relevant information comprising at least the steps of:
  • search query containing at least one search term
  • At least one logical construction of said transformed structured source data array is identified, comprising at least said grammatically and spelling-correct semantic part;
  • At least grammatically and spelling-correct semantic parts associated with said grammatically and spelling-correct semantic parts are contained in other logical constructions of the transformed structured data array, said identification being carried out based on information from the component map of grammar and spelling true semantic parts of logical sections of logical constructions;
  • FIG. 1 by way of example, but not limitation, a general flowchart of the steps of the claimed method 100 for transforming a structured source data array containing at least natural language text is depicted.
  • the claimed method 100 for converting a structured source data array containing at least natural language text is characterized by performing step 101 of generating an initial data structure of a structured source data array on which an initial data structure is formed a structured source data array containing the source elements of the original data structure, said source elements being structurally complex language constructs consisting of structurally complex language constructs containing contextual terms and structurally complex language constructs not containing contextual terms, and structurally -simple language constructs containing structurally simple language constructs containing contextual terms and structurally simple language constructs not containing contextual terms.
  • step 101 of generating the initial data structure of the structured source data array an analysis of the elements of the structured source data array is performed to determine if it contains the source elements of the original data structure of the structured source data array.
  • the initial data structure of the structured initial data array is formed, which consists of structurally complex and structurally simple language structures - sentences and enumeration headings.
  • the identification of structurally complex language structures is carried out in accordance with the first identification criterion, which is an identification criterion for the first type of combination of punctuation marks.
  • the first type of combination of punctuation marks is characterized by the presence of the following features of a structurally complex language structure:
  • each of the second text fragments begins with a capital letter
  • each of the second text fragments can begin with a list marker - a formatting character (typographic sign) used to highlight list items (listing headings);
  • each of the second text fragments may begin with a numbering marker - numbers or letters with a parenthesis or period;
  • the number of second text fragments located on the red lines of paragraphs must be at least two.
  • structurally simple language constructs are those elements of the initial data structure in which the combination of punctuation marks does not meet the first criterion, i.e. such elements of the original data structure satisfy the identification criterion for the second type of punctuation mark combination, the second type of punctuation mark combination being all punctuation mark combinations, except for punctuation mark combinations of the first type.
  • the identification of structurally simple linguistic constructions is carried out by identifying in the natural language text the signs of the end of the sentence - punctuation marks “.” (“Dot”), “...” (“ellipsis”), etc., in conjunction with signs the beginning of the next sentence - in capital letters, numbers, etc., taking into account the presence and certain combinations of punctuation marks, namely punctuation marks (“.” (“dot”), “;” (“semicolon”), “ ... ”(“ ellipsis ”),“, ”(“ comma ”),“ (”(“ opening bracket ”),“) ”(“ closing bracket ”),“: ”(“ colon ”), etc. p.), with water separators (“” (“space”), “-” (“dash”), etc.) and typography ( ⁇ > (“paragraph”), number, “°” (“degree”), etc. .).
  • the method 100 for converting a structured source data array is characterized by performing a step 102 of identifying context terms and forming a database of context terms, where context terms are identified in said source elements and forming a context terms database containing said identified context terms.
  • combinations of contextual terms of the first type, contextual terms of the second type, contextual terms of the third type, which as a result are identified as contextual concepts, contextually generalized concepts and contextual links, are identified. Identification by the criterion of a combination of contextual terms of the first type allows the identification of contextual concepts.
  • Contextual concepts are a one-word or verbose term that does not have an independent semantic meaning, the true meaning of which is clarified with the help of another, synonymous or associative term (group of terms), located in the same sentence as the contextual concept, or in previous sentences. Identification by the criterion of a combination of contextual terms of the second type allows the identification of contextually generalized concepts.
  • Contextually generalized concepts are single-word or verbose terms, one of which is a generalizing word, and the other generalized words, while the term being a generalizing word includes the lexical meaning of a number of homogeneous members (generalized words) and answering the same question as homogeneous members.
  • Contextual links are a verbose term that does not have an independent semantic meaning, which explicitly or implicitly refers to other sentences that contain the semantic meaning of this term.
  • contextual link does not indicate the exact location of the meaning, then such a link is an implicit contextual link, and the “degree of implicitness” of the implicit contextual link is inversely proportional to the value of the “degree of accuracy” of the location of the meaning.
  • An example of an implicit contextual link with a “maximum degree of implicitity” is a contextual link of the form “specified requirements”. The identification of contextual terms reveals, in fact, contextual terms and terms that clarify their semantic meaning. Between contextual terms and terms that clarify their semantic meaning, a contextual relationship is recorded, which is the reason for the formation of the corresponding entries in the database of contextual terms.
  • Contextual terms identified in sentences are not necessarily removed from sentences and may remain in them, but their meaning will be clear and unambiguous, as this meaning will be clarified in the relevant contextual reference books (reference book of contextual concepts, reference book of contextually generalized concepts, reference book of contextual links).
  • a method 100 of transforming a structured source data array characterized by the execution of the step 103 of identifying the elements of the original data structure, on which, using the information contained in said generated database of contextual terms, structurally complex language constructs containing contextual terms, structurally complex language constructs not containing contextual ones are identified in said initial data structure terms, structurally simple language constructs containing contextual terms and structurally simple language constructs not containing cont kstnye terms.
  • the initial elements of the initial data structure of the structured initial data array are identified.
  • an unambiguous identification of the source elements and their grouping is carried out according to the signs of the presence or absence of contextual terms in them and the type of the source element.
  • structurally complex linguistic constructs containing contextual terms, structurally complex linguistic constructs not containing contextual terms, structurally simple linguistic constructs containing contextual terms, and structurally simple linguistic constructs not containing contextual terms are uniquely identified.
  • the method 100 for converting a structured source data array is characterized by performing a first type conversion step 104, where the first type is converted over said structurally complex language constructs that do not contain contextual terms to obtain grammatically and spelling-correct structurally simple language constructs that do not contain contextual terms.
  • Structurally complex linguistic constructions that do not contain contextual terms are subjected to transformations with the aim of transforming enumeration headings into sentences.
  • structurally complex linguistic constructions that do not contain contextual terms are converted into structurally simple linguistic constructions that are sentences in the usual sense of the term. This conversion is initially formatted in nature, i.e. non-formatted text of a structurally complex linguistic structure is converted into formatted, i.e.
  • linguistic transformations are carried out. Linguistic transformations over created structurally simple language constructions associated with the restoration of correct grammar and spelling.
  • the indicated linguistic transformations are understood as: a) the coordination of sorts, numbers, cases; b) editing (replacement and removal) of inappropriate punctuation marks. All the above transformations performed on structurally complex language constructs that do not contain contextual terms are transformations of the first type.
  • the method 100 for converting a structured source data array is characterized by performing a second type conversion step 105, in which the second type is converted over said structurally simple language constructs containing contextual terms to obtain grammatically and spelling-correct structurally simple language constructs that do not contain contextual terms.
  • Structurally simple language constructs containing contextual terms are transformed in order to structure and clarify the semantic content of contextual terms.
  • structurally simple linguistic constructions containing contextual terms are transformed into grammatically and spelling-correct structurally simple linguistic constructions that do not contain contextual terms. This transformation is structurally linguistic in nature, i.e.
  • the mentioned transformation may be associated with the addition of the text of a structurally simple language construct containing contextual terms, the formation of a new entry in the corresponding contextual reference.
  • Each type of contextual term applies its own type of transformation. So for contextual concepts, the first transformation of the second type is used, for contextually generalized concepts - the second transformation of the second type, for contextual references - the third transformation of the second type.
  • a linguistic conversion is carried out to eliminate grammatical and spelling errors associated with the conversion of structurally simple language constructs containing contextual terms into structurally simple language constructs that do not contain contextual terms.
  • Linguistic transformations over created structurally simple language constructs that do not contain contextual terms are associated with the restoration of correct grammar and spelling.
  • Under specified linguistic transformations are understood: a) the coordination of sorts, numbers, cases; b) editing (replacement and removal) of inappropriate punctuation marks. All of the above transformations are transformations of the second type.
  • the first transformations of the second type are carried out over contextual concepts and revealed terms that clarify the semantic meaning of contextual concepts. These identified terms can be either contextual synonyms or contextual associations.
  • a context synonym is a synonym that can be used as a synonym only in this context.
  • the contextual term “her” is, in fact, a contextual concept
  • the identified term “organization” is its unintelligent synonym.
  • the first transformation of the second type provides the formation of a directory of contextual concepts, which, in turn, provides clarity of the semantic meaning of contextual concepts.
  • linguistic text correction is provided by replacing contextual concepts with a contextual synonym or contextual association.
  • the sentence: “The manufacturer is an organization regardless of its legal form” through the implementation of the first transformation of the second type on it will be transformed into the sentence: “The manufacturer is an organization regardless of the legal form of organization”.
  • Contextual associations differ from contextual synonyms in that they, as a rule, represent a more complex textual construction, consist of several terms and are rarely found in one sentence with a unified concept, but more often in other sentences.
  • the second transformations of the second type are carried out over contextually generalized concepts.
  • context-generalized concepts mean a generalizing word (generalizing term) and homogeneous words (homogeneous terms), as well as a group of homogeneous words (a group of words without a generalizing word).
  • a generalizing word is a word or phrase that is common to homogeneous words or phrases located near the generalizing word.
  • homogeneous words in the context of the present invention are those members of the sentence that depend on the same word and answer the same question (i.e. they are homogeneous members of the sentence).
  • a generalizing word is the same type of sentence members as words or phrases homogeneous with it. A significant part of the proposals complicated by homogeneous members can be presented as a result of the “compositional reduction” of a number of independent proposals. Second conversion of the second type ensures the removal of all terms that are homogeneous with the generalizing word from the analyzed sentence, due to which only the term that is the generalizing term remains in the sentence.
  • a directory of contextually generalized concepts is formed, which contains the identified generalizing term and all the homogeneous words corresponding to it, generalized words (generalized concepts).
  • the second transformation of the second type provides a linguistic adjustment of the sentence by creating new sentences that use generalized concepts instead of a generalizing term. For example, in the sentence: “The Contractor is an organization that performs work or provides services to consumers under a reimbursable contract”, there are homogeneous words (homogeneous terms) - the term “performing work” and the term “providing services”.
  • the method 100 for converting a structured source data array is characterized by performing a third type conversion step 106, in which the third type is converted over said structurally complex language constructs containing contextual terms to obtain grammatically and spelling-correct structurally simple language constructs that do not contain contextual terms.
  • a transformation of the third type is a combination of transformations of the first type and transformations of the second type, applied to structurally complex linguistic constructions containing contextual terms.
  • the conversion of the third type is a two-stage conversion. At the first stage, structurally complex linguistic constructions containing contextual terms are converted to the first type, due to which the structurally complex linguistic construct containing contextual terms is converted to a structurally simple linguistic construct containing contextual terms.
  • the method 100 for converting a structured source data array is characterized by performing step 107 of forming a final data structure of a structured source data array, on which a final data structure of a structured source data array containing the elements of the final data structure is formed, the elements of the final data structure being structurally simple language constructions that do not contain contextual terms and obtained through transformations of the first, second, and third th types grammatically and spelling-right structurally simple language constructs that do not contain contextual terms.
  • the resulting data structure is formed from.
  • structurally simple language constructs that do not contain contextual terms structurally simple language constructs that do not contain contextual terms
  • structurally simple language constructs that do not contain contextual terms obtained by converting structurally complex language constructs that do not contain contextual terms
  • structurally simple linguistic constructions that do not contain contextual terms obtained by converting structurally complex linguistic constructions containing contextual terms.
  • the final data structure consists of elements - grammatically and spelling-correct structurally simple language constructs that do not contain contextual terms, which from a linguistic point of view are sentences that do not contain contextual terms.
  • FIG. 2 depicts a General diagram of the steps of the claimed method 200 for converting a structured source data array containing at least natural language text.
  • the claimed method 200 for converting a structured source data array containing at least natural language text is further characterized by the execution of step 201 of generating the first data structure, based on which obtained on the basis of formation 107 the final data structure of the structured source data array form the first data structure of the structured data array containing the elements yanutoy first data structure, said elements of the first data structures representing the final elements of the data structure of a structured data source array comprise first and second logical partitions are logical partitions; performing step 202 of forming a database of logical connections of logical partitions, on which a database of logical connections of logical partitions of said elements of the first data structure is formed; performing step 203 of generating a second data structure, which forms a second data structure of a structured data array containing elements
  • step 205 the formation of grammatically and spelling-correct semantic parts, on which grammatically and spelling-correct semantic parts of said logical sections are formed by linguistic transformations over said semantic parts; and performing step 206 of generating the resulting data structure, which forms the resulting data structure of the structured data array containing the elements of said resultant data structure, said elements of the resulting data structure containing logical constructs containing at least the grammatically and spelling-correct semantic parts of the logical sections.
  • Step 201 is characterized by performing step 2011 of identifying the source structure, in which the resulting data structure 1 of the structured data array is identified; performing element identifying step 2012, wherein elements 11 of the final data structure 1 are identified; the implementation of the stage 2013 identification of logical partitions, which identify the first logical partitions 111 of the elements 11 of the final data structure 1 and the second logical sections 112 of the elements 11 of the final data structure 1; and the implementation of the stage 2013 of the formation of the first data structure, which forms the first data structure 2 of the structured source data array containing the elements 21 of the first data structure, which are the elements 11 of the final data structure 1, the elements 21 of the first data structure 2 containing the first logical sections 111 elements 11 of the final data structure 1, and containing elements 22 of the first data structure, and elements 22 of the first data structure 2 contain second logical sections 112, elements 11 of the final structure 1 data.
  • the final data structure 1 is the final structure obtained at step 107 a data structure, which, in turn, is a data structure of a structured source data array containing at least natural language text.
  • a data structure which, in turn, is a data structure of a structured source data array containing at least natural language text.
  • NLA regulatory legal act
  • the resulting data structure 1 contains elements 11, which are positions that are sentences - grammatically organized word combinations.
  • each sentence is characterized by semantic completeness, because is an element of the final data structure obtained at step 107.
  • the identification of proposals at the stage of 2012 identification of elements is carried out by identifying in the natural language text signs of the end of the sentence.
  • Signs of the end of the sentence are: period, semicolon, ellipsis, etc.
  • Identification of proposals is carried out in conjunction with the identification of signs of the beginning of the proposal. Signs of the beginning of a sentence are: a capital letter, a number, a number with a closing bracket, a number with a period, etc.
  • the identification also takes into account the presence of certain combinations of punctuation marks, namely punctuation marks - periods, semicolons, brackets, colons, etc., - word separators, namely space, etc., and a typography - paragraph. numbers, degrees, etc.
  • Punuation marks namely punctuation marks - periods, semicolons, brackets, colons, etc.
  • - word separators namely space, etc.
  • elementary semantic units are identified - sentences that are judgments.
  • a simple judgment is a judgment, no part of which is a judgment.
  • the linguistic form of expression of judgment is narrative sentences.
  • the identification of elements 11 of the final data structure 1 is carried out by identifying and defragmenting the sentence into the primary components of the sentence, namely into words, particles, conjunctions, prepositions, etc., and punctuation marks.
  • concepts expressed in separate words and / or phrases based on various directories and dictionaries are formed from the primary elements.
  • simple judgments are formed from the formed concepts, which are groups of interconnected concepts, while the interconnectedness of concepts is determined on the basis of syntactic or other relationships between the concepts.
  • a linguistic-semantic analysis of elements 11 of the final data structure 1 is carried out, whereby elements 11 of simple judgments are identified in elements 11.
  • the structural elements of simple judgments are understood as the subject of judgment, the predicate of judgment, the connective and the quantifier word.
  • the subject of the judgment (S) is the concept expressing the subject of the judgment, that is, what is said in this judgment.
  • the predicate of judgment (P) is a concept that expresses this or that information about the subject of judgment.
  • Subject of judgment and the predicate of judgment are the basic structural elements of judgment, which are terms of judgment.
  • the connection between the subject of judgment and the predicate of judgment which reflects the real relationship between objects conceivable in concepts, is revealed through a logical connective.
  • connective In Russian, the connective is expressed by the words: “is” (“is not”), “is” (“is not”), “is” (“is not available”), etc., is indicated by a dash, a colon, and can also be implied by the agreement of words (“It is raining”, “The dog barks”).
  • a connective is a logical constant, because it contains unchanged content - it always serves as an indicator of the presence or absence of something in the subject of thought.
  • a quantifier word (for example, “everyone”, “all”, “none”, “some”, etc.) indicates whether the information on the predicate of a judgment applies to the entire volume of a concept expressing a subject of a judgment, or to its part.
  • the identification is based on the results of the identification of the identified judgments, defined as complex judgments.
  • Complex judgments are a group of simple judgments in which there is a connection between separate judgments, established using logical conjunctions "and”, “or”, “if ... then ", “then and only then ..., when ",” it is not true that ... ".
  • Types of relationships between individual judgments are expressed by the corresponding logical connectives and are shown in Table I.
  • connection is determined by the meaning of logical alliances, which consists in the answer to the question: “Under what conditions will a complex judgment be true, and under what conditions will it be false?” In other words, under what combinations of truth and falsity of simple judgments that make up a complex proposition, a logical union determines the true connection, and in which - the false one. A judgment is considered true if the description it gives is true (a real situation), and false if it does not. “Truth” and “false” are called truth values of judgment and are the main logical characteristic of judgments.
  • conditional proposition is a complex proposition in which simple propositions are united by a logical union “if ... then ...”. For example: “If a citizen violates the law, this creates liability for the violation” or “If the number is divisible by 2 without a remainder, then it is even.”
  • a conditional proposition consists of two types of judgments that comprise it. The judgment written after the word “if” is called the basis (previous).
  • conditional propositional formula can be represented as “A—> B”, where A is the basis, B is the consequence.
  • A is the basis
  • B is the consequence.
  • the foundations and consequences in themselves, can be both simple judgments and complex judgments. Formed from the previous and subsequent judgments, a conditional proposition, first of all, implies that it cannot be that what the foundation says has place, and what the investigation says is missing. In other words, if the ground is true and the effect is false, then such a conditional proposition will be false. This condition determines that a conditional proposition is true in all cases except one: when the previous one is, and the next is not, i.e.
  • the first data structure 2 formed in step 201 contains elements such as sentences (element 11 of the original data structure 1) and judgments (logically sections 111 or 112 elements of source data structures 11 1). Moreover, judgments are additionally identified by the mentioned logical connection as grounds, i.e.
  • the first logical section 111 of the element of the initial data structure 11 which has a logical connection of the 1st type and is the judgment “A”
  • the second logical section 112 of element 11 of the original data structure 1 having a logical connection of the 1st type and which is the judgment “B”.
  • all elements of the original data structure 1 are separated by the presence of the aforementioned first logical partitions 111 or second logical partitions 112 of the original data structure 1, whereby the elements 21 of the first data structure 2 are formed having the first logical partitions 11 1 and elements 22 of the first data structure 2 having second logical partitions 112.
  • step 202 of creating a database 3 of logical connections of logical partitions is characterized by the execution of step 2021 of identifying the elements of the first data structure 2, which identifies the elements 21 of the first data structure 2 containing one of the first logical partitions 111, which are the elements 31 of the first data structure 2, and the elements 22 of a first data structure 2, comprising one of said second logical partitions 112, which are elements 32 of a first data structure 2; performing step 2022 of identifying the elements of the first data structure 2, which identifies the elements 21 of the first data structure containing more than one of the first logical partitions 111, which are the elements 33 of the first data structure 2 and the elements 22 of the first data structure containing more than one of the second logical partitions representing the elements 34 of the first data structure 2; the execution of step 2023 identification of logical relationships, in
  • all elements of the first data structure 2 namely arrays of elements of 21 sentences containing the first logical sections 111 and elements of 22 sentences containing the second logical sections 112, must be separated into groups of elements 31, 33 and 32, 34, containing either only the first logical partitions 111, or only the second logical partitions 112, respectively.
  • each element included in the arrays of elements 31, 33 containing the first logical partition 111 is identified as an element 31 having only one logical partition 111, or as an element 33 having more than one logical partition
  • the identified logical elements 32, 34 have the first logical partitions 111, the first logical partitions 111 are deleted from the identified elements 32, 34.
  • the resulting arrays of elements 32, 34 they are also still connected with the element from which they are selected and for this reason are identified as separate elements of this array of elements.
  • the nature of the logical relationships between the same type of judgments in the elements of two created arrays of elements 31, 33 and the elements 32, 34 is established.
  • a conjunctive (connective) proposition is a complex proposition formed from the initial propositions by means of a logical union “and”, denoted by the symbol “ l ”.
  • a L B This complex proposition can be represented by the formula: “A L B”, where A, B are conjuncture elements; " L " - a symbol of logical union - conjunction.
  • the conjunctive logical union is expressed by many grammatical unions: “and”, “a”, “but”, “yes”, “although”, “however”, “and also ...”. Often such grammatical unions are replaced by punctuation marks - a comma, a colon, a semicolon.
  • a disjunctive (dividing) proposition is a complex proposition formed from "initial" propositions by means of a logical union "or”, denoted by the symbol "V”.
  • the proposition: “The law can contribute to economic development or hinder it,” is a disjunctive proposition consisting of two simple judgments: “The law can contribute to economic development” and “The law can impede economic development”. Accordingly, having designated them through the letters A, B, such a judgment can be represented through the formula: “A V B”. Since the connective "or” is used in two different meanings - non-exclusive and exclusive, they distinguish between weak and strong (strict) disjunctions. A weak disjunction is true when at least one of its component judgments (or both together) is true and false when both its judgments are false (see table 2, column 4). A strong disjunction (the symbol “W”) differs from a weak disjunction in that its components are mutually exclusive.
  • a denied proposition is a complex proposition formed by the logical union “it is not true that ...” (or simply “not”), which is usually represented by a negative sign (“-” symbol). Unlike the binary unions mentioned above, such an alliance refers to one proposition.
  • Adding this union to a proposition means the formation of a new proposition, which depends on the original proposition - the denied proposition is true if the original proposition is false, and vice versa (see table 2, columns 8, 9). For example, if the initial judgment is: “All witnesses are truthful,” then the denied: “It is not true that all witnesses are truthful.” If a separate logical section (simple proposition) remains unidentified from the point of view of the logical nature of the proposition, then as a result of linguistic-semantic analysis of the text surrounding the sentence containing such a section, the actual contextual form of this simple proposition can be revealed.
  • the first data structure contains logical sections of sentences that form the original data structure.
  • Step 203 of the formation of the second data structure 4 is characterized by the execution of logic construction step 2031, in which logical structures 41 of logical partitions 111, 112 of elements 31 are formed , 32, 33, 34 of the first data structure 2, using information from the database 3 of logical connections of the logical partitions of the elements 31, 32, 33, 34 of the first data structure 2 and the logical partitions of the mentioned elements 31 of the first page data structures 2 containing one of said first logical partitions 111 and logical partitions of said elements 32 of a first data structure 2 containing one of said second logical partitions 112; and performing step 2032 of generating a second data structure 4, wherein a second data structure 4 is formed comprising elements 41 of the second data structure 4, said elements of the second data structure 4 being formed logical structures 41 of logical sections 111, 112 of elements 31, 32, 33 , 34 of the first data
  • Logical constructions 41 are the result of data conversion convertible structured data array. Logical constructions 41 are formed in accordance with the specifics of the converted text in natural language, in particular, normative documents.
  • the specificity of the legal acts is that it contains the rule of law (legal norms).
  • the specifics of the legal acts are that in the theory of the rule of law there are concepts of a logical rule of law and a legal rule of law. These concepts are not identical.
  • the logical rule of law includes the content of all elements of the rule of law established in legal science, including hypothesis, disposition and sanction, and the legal rule of law reflects specific * regulatory requirements contained in specific proposals of specific legal acts.
  • the difference lies in the fact that one particular logical rule of law can be contained in a specific set of legal rules of law, i.e. in a multitude of regulatory requirements.
  • the logical design is the basis (framework) of the main regulatory requirement, containing two basic structural elements - “situation” and “rule”.
  • the main regulatory requirement (hereinafter referred to as the regulatory requirement) is an instrument of legal regulation and includes regulatory and protective regulatory requirements.
  • the situation in the regulatory requirements means any conditionality of the rule
  • the rule means any rules, including the rules (model) of behavior of subjects of legal relations.
  • a situation is a judgment having a logical implicative connection and being a basis
  • a rule is a judgment having a logical implicative connection and being a consequence.
  • logical structures 41 i.e. It is also necessary to take into account regulatory requirements that each of the elements of this design (both the situation and the rule) can consist of both a single judgment and a group of judgments.
  • logical constructions it is necessary to use the database of logical connections of logical partitions. In addition to the identified logical connections between logical sections for the formation of a logical structure, you must refer to the rules for the formation of logical structures.
  • the rules for the formation of logical constructions reflect the requirements of legal science and legal practice in relation to the composition and structure of a normative prescription (prescription). For example, the condition that one prescription cannot contain two different rules leads to the fact that the rules establish that if one sentence contains two consequences that have a logically weak disjunctive connection, this means that these judgments are different rules and accordingly different regulations. Moreover, if the same two consequences have a logically strong disjunctive connection, then this unites them in the framework of one a complex rule within one prescription. In essence, the rules for the formation of logical constructions 41 are reduced to permissible combinations of logical connections between the same type of judgments within the framework of a single normative prescription.
  • the second data structure formed in the above manner contains such elements as judgments (logical section of the element of the original data structure) and regulatory requirements (logical structure 41 of logical sections of the element of the original data structure) (Fig. 8). In this case, judgments are identified by the presence of an implicative logical connection into two main logical sections:
  • foundations (the first logical section of the element of the original data structure containing a logical implicative connection, a connection of the 1st type, type A); 2) consequences (the second logical section of the element of the original data structure, containing a logical implicative connection, a connection of the 1st type, type B);
  • FIG. 9 depicts a General diagram of the steps of step 204 of the formation of the database 5 semantic parts.
  • Stage 204 of the formation of the database of 5 semantic parts is characterized by the execution of step 2041 identification of logical partitions, which identify the first logical partitions 411 elements 41 of the second data structure 4 and the second logical partitions 412 elements of the second data structure 4; by performing a semantic part identification step 2042, wherein in the first logical partitions 411 and second logical partitions 412 of the elements of the second data structure 4, the first semantic parts 4110 and the second semantic parts 4120 are identified; and by performing step 2043 of identifying specific semantic parts, in which at least the specific semantic parts 4111 of the first logical sections 411 of elements 41 of the second data structure 4 and special semantic are identified in the first and second logical sections 411, 412 of the elements 41 of the second data structure parts 4121 of the second logical partitions 412 of the elements 41 of the second data structure 4 and form a database 5 of special semantic parts of
  • the logical structures 41 formed in the second data structure are the framework and the basis of the normative prescription, but still do not fully comply with it.
  • To achieve maximum compliance with the structure of logical constructions 41 to the structure of a normative prescription it is necessary to conduct a comprehensive semantic analysis of logical sections 411, 412, including at least syntactic and logical analysis of terms and concepts, identifying the relationships between concepts of judgment and between terms of complex concepts.
  • the purpose of this semantic analysis is to identify and identify in logical sections 411, 412 logical structures 41 of the second data structure 4 rows of specific parts (second parts) of logical sections that lead to:
  • the identification of specific parts i.e. the identification in logical sections 411, 412 of logical structures 41 of the second data structure of the first and second semantic parts 4110, 4120 of logical sections 411, 412.
  • the first semantic parts! 110 are formed by removing from the logical sections 411, 412 the second semantic parts 4120 (specific parts).
  • the first semantic part 4110 of the logical section is the semantic core of judgment, i.e. judgment cleared of specific parts.
  • the semantic core of a judgment is the basic elements of a judgment, such as the subject of a judgment, the predicate of a proposition, and a connective discloses the inclusion (or exclusion) of a subclass in an object class or the belonging (non-belonging) of an element to a class.
  • a judgment “Crime is an unlawful act”
  • the subject of the judgment is the word “crime”
  • the predicate of the judgment is the phrase “unlawful act”
  • the link is the word “is”.
  • the second semantic parts of the 4120 logical section are the concepts of judgment, which are identified as signs of the subject of the judgment, the predicate of judgment, as well as the terms of the judgment — a connective (when it can be interpreted in the “three-dimensional plane”) and a quantifier word, as well as other, special parts.
  • the concept (subject of judgment) “crime provided for by the criminal Code” contains the concept - the word “crime” and the sign of the concept - the phrase “provided for by the criminal Code”.
  • Signs of the concept are the content of the concept, indicating the presence or absence of one or another property, state or relationship. In other words, a sign of a concept is all that in which the concepts can be similar or different from each other.
  • a quantifier word indicates whether information about a predicate of a judgment applies to the whole volume of a concept expressing a subject of a judgment, or to a part of it. For example, in the judgment: “Every crime is an unlawful act”, the quantifier word “any” indicates that information about the subject of the judgment (the phrase “unlawful act”) refers to the entire volume (to each element of the volume) of the subject of the judgment - the word “crime” .
  • step 205 of forming grammatically and spelling correct semantic parts is depicted, on which grammatically and spelling correct semantic parts of said logical sections 41 are formed by linguistic transformations over said semantic parts.
  • Step 205 includes performing a refinement structure identification step 2051, wherein at least second refinement structures of the second semantic 4122 parts of the second logical partitions 412 are identified in said second semantic parts 4120 of said second logical partitions 412 of elements of the second data structure 4; and performing a step 2052 of linguistic transformations, where linguistic transformations are performed on all semantic parts, with the exception of the mentioned special semantic parts 4111, 4121 of the first and second logical sections 411, 412, to form grammatically and spelling-correct semantic parts 4123 and qualifying logical structures 4122 sections of elements 41 of the second data structure 4.
  • a general diagram of the obtained second data structure 4 is shown in FIG. 12.
  • the subject of analysis will be an array of values identified by the indicated types of the second semantic parts 4120, i.e. an array of concepts contained in logical sections and identified as corresponding species. Each concept of these arrays must be identified in terms of its belonging to refinements 4124 or to dependencies 4125. Moreover, refinement is such a characteristic of a concept that carries out the transition from a broader concept to a narrower one, and the dependencies contain signs of a legal fact, i.e.
  • FIG. 13 by way of example, but not limitation, a general flowchart is shown of the steps of step 206 of generating the final data structure 6, in which the resulting data structure 6 of the structured source data array is formed containing the elements 61 of said resultant data structure 6, the aforementioned elements 61 of the final data structure contain logical constructions 61 containing at least the grammatically and spelling-correct semantic parts 4123 of logical sections.
  • Step 206 is characterized by the execution of a semantic combination step 2061, in which 4123 second logical partitions 4 of the second data structure 4 and 4 grammatical and spelling correcting structures 4122 of the second semantic parts 4120 of the second logical partitions 412 of the elements are formed from the first grammatically and spelling-correct semantic parts 4123 41 second data structures 4 semantic combinations 611 grammatically and spelling-correct semantic parts 4122, 4123 second logical sections 412 element o41 of the second data structure 4; and by performing step 2062 of generating the resulting data structure 6, wherein the resulting data structure 6 is formed comprising the elements 61 of the resulting data structure 6, said elements 61 of the resulting data structure 6 being logical structures 61 containing the grammatically and spelling-correct semantic parts 4122, 4123 logical partitions of the elements 41 of the second data structure 4.
  • the final overall structure of element 61 of the resulting data structure 6 is shown in FIG. 15.
  • the basis of a normative prescription is a legal rule (rule), which can be formed as correctly as possible as a result of a comprehensive semantic analysis of logical sections of logical structures. At the stage of formation of logical constructions, most of the conditions were separated from the rule (second logical section 412) and separated into a separate section - the first logical section 411.
  • the semantic core of the rule (second logical section 412) was revealed and the rest of the conditions were they are embodied in the second semantic parts 4120, in which some of these parts are identified as refinement 4124.
  • FIR 61 structures create a semantic coupling 611, i.e. combinations of the first semantic parts 4110 of the second logical sections 412 and the second semantic parts 4120 of the second logical sections 412, identified as refinement 4124.
  • These semantic combinations are legal rules.
  • the resulting data structure is a structured structure, the elements of which are most consistent with the structure of the regulatory prescription.
  • the resulting data structure is formed in this form in order to simplify and regulate the professional work on the creation and adjustment of regulatory documents.
  • the resulting data structure is a design that allows you to literally visualize regulatory requirements, see all the actual elements of the semantic design, which allows them to be comprehensively analyzed in order to make precise adjustments to both existing regulatory requirements and draft regulations at different stages of their creation.
  • FIG. 16 depicts a General diagram of the steps of the claimed method 300 of forming a map of the relations of the components of the grammatically and spelling correct semantic parts of the logical sections of the logical structures of the converted structured source data array.
  • the claimed method 300 for generating a map of component relationships is characterized by performing a step 301 of identifying components and values, which identify the components of the grammatically and spelling correct semantic parts of logical sections of logical constructions from said transformed structured source data array, each grammatically and spelling-correct semantic part containing said components, contains at least one said component, wherein said component contains no more than one component value, and form a component table containing at least said identified components and their values; step 302 of identification of semantic parts, which identify grammatically and spelling correct semantic parts of logical sections of logical structures containing said identified components, and grammatically and spelling correct semantic parts of logical sections of logical structures that do not contain said identified components; component identification step 303, in which components containing more than one in said grammatically and spelling correct semantic parts are identified and combinations of components contained in more than one in each separate grammatically and spelling correct semantic parts are identified, and a map of component combinations containing at least the aforementioned components contained in an amount of more than one in each separate gramm
  • the method 300 of generating a map of the connections of the components may include a step 310 of generating a report on which a report is generated showing at least the number of links and the values of said components contained in the map of combinations of components.
  • a component is an attribute of a legal rule.
  • a legal rule is part of a regulatory enactment.
  • the purpose of the component structuring of rules is the formation of a certain design standard and systematic technical assessment of the construction of a legal rule, as well as the search and identification of related requirements that form the construction of a logical rule of law. All components of the rule mentioned in the present description are examples used for the purpose of explaining the present invention. Other components without restrictions may be identified in accordance with this method in other legal rules.
  • the first component of a legal rule is the subjects of legal relations - these are authorized persons established in the precept, relations between which are governed by the precept.
  • one of the parties for which a certain rule is established may be used in the prescription.
  • Parties to legal relations are referred to as the subject of the order and the counter-subject.
  • the subject of the prescription is the cause and source of regulatory influence
  • the counter-subject is the party to the legal relationship with which this impact.
  • the second component of a legal rule is the regulatory impact - an action, inaction, requirement, principle, etc.
  • the regulatory impact will be the content of the legal relationship - i.e. the possibility of certain actions, the need for certain actions or the need to refrain from prohibited actions. If the order does not contain the subject of legal relations, then the regulatory impact will be the content of the requirements relating to the objects of legal relations.
  • the third component of a legal rule is the object of legal relations - these are objects, phenomena and their derivatives (hereinafter referred to as objects), these are objects of the world around which the regulatory influence is directed. If the rule does not contain the subject of legal relations, then the objects of legal relations are those objects that are subject to regulatory influence in the form of the content of requirements for these objects.
  • the fourth component of a legal rule is related objects - these are objects, phenomena and their derivatives that are present in the rule, but are not objects of legal relations. Related objects are used in the rule either to detail and clarify the regulatory impact, or as an object of regulatory impact, or as an object of regulation.
  • Each component (type of component) installed in the component structure is described using various linguistic tools, such as syntactic functions and relationships, parts of speech, sentence members, etc.
  • Linguistic tools describe individual terms (concepts) and groups of terms (concepts) included in the semantic parts. Such descriptions are formed in the form of logical formulas in which the elements are the mentioned linguistic tools connected with each other by logical connectives - conjunction ("and"), disjunction ("or”), strict disjunction ("either, ... or”), implication (“if .. ., then ").
  • Logical formulas for each component (type of component) form a database of descriptions of components.
  • Component descriptions database is a set of tools for finding components in grammatically and spelling-correct semantic parts of logical sections of logical constructions of a transformed structured source data array.
  • the map of component relationships in this case consists of internal and external component relationships.
  • the internal relationships of components indicate combinations of components within one rule (within the semantic part containing components).
  • External relations of components indicate the relationship between individual regulatory requirements based on the presence of semantic relations between components and individual words, and phrases in semantic parts that do not contain components.
  • the components themselves are identified.
  • the identification of components is meant the identification in each separate semantic part containing the components of the following information: a) the name of the component; b) the type of component; c) the value of the component.
  • the name of the component is an element of a legal rule, reflecting the signs of legal relations in a particular rule, which is the main part of the linguistic formula.
  • the type of component is the main and additional parts of the linguistic formula, i.e. separate subgroups of component names identified on the basis of differences in the additional part of the linguistic formula.
  • the meaning of a component is a term or group of terms contained in the semantic part that correspond to the main and additional parts of the linguistic formula. Table v
  • the components will be identified in the legal rule: "The authorized body is obliged to take measures to recall such goods from the domestic market from the consumer.” In this example, all five components of the rule are identified (see table VI).
  • each component can have one value.
  • a table of components contains the following information: name of the component; type of component; component value unique semantic part number; Unique logical partition number unique logical design number.
  • the semantic parts containing more than one component and semantic parts containing one component are identified. The reason for this separation of semantic parts containing components is that the component combination map is an array of actual component combinations in each individual rule, i.e. in each separate semantic part containing the components.
  • a map of component combinations can be formed on the basis of semantic parts containing more than one component.
  • each semantic part containing more than one component forms a group of components of one rule.
  • Component combinations within each individual component group form a map of component combinations.
  • the component combination map contains the following information: name of the combined components; types of components to be combined; meaning of combined components, unique number of the semantic part; Unique logical partition number unique logical design number.
  • cards of combinations of components in the legal rule will be produced: “The authorized body is obliged to take measures to recall such goods from the domestic market from the consumer”.
  • An example of an incomplete coincidence is a selection of prescriptions in which all prescriptions contain a part of the components specified by the user from the specified combination — for example, only the combination K1-KZ.
  • Such a selection includes not only prescriptions with the established combination of K1-KZ, but also with other combinations of components (for example, K1-K2-KZ), in which, along with the components indicated in the combination K1-KZ, contain other components (for example, K2)
  • Map component combinations is the first part of a component association map.
  • the external relations of the components that make up the second part of the map of component relationships are based on the presence of semantic relationships between components and individual words, and phrases in semantic parts that do not contain components.
  • the semantic connection is understood as the presence between the meanings of the components and individual words, and phrases of semantic coincidence, i.e. coincidence of the meanings of individual components and words within the meaning.
  • one rule can be considered (the semantic part containing the components) - “The seller must transfer the goods to the consumer”, and one semantic part not containing the components (situation, legal fact) - “If the goods are not transferred to the consumer ".
  • it is necessary to identify the external relations of the components, i.e. the presence of a semantic connection between the meanings of components and concepts that semantically coincide with them.
  • semantically coinciding concepts can be not only completely matching or one-root words (terms), but also their synonyms and associations identified in special connected dictionaries and reference books. All concepts that semantically coincide with the values of the components identified in this way are recorded in a special table - a map of combinations of semantically coinciding concepts.
  • a map of combinations of semantically coinciding concepts establishes relationships between individual semantic parts containing components and semantic parts that do not contain components. This relationship is based on identified relationships between components and related concepts. As a result, related requirements are identified.
  • a map of combinations of semantically matching concepts contains the following information for each pair of related prescriptions: name of components; types of components to be combined; the value of the combined components; unique number of the semantic part containing the components; unique number of the logical section containing the mentioned semantic part; a unique logical construction number containing the logical partition; the name of concepts that semantically coincide with the meanings of the mentioned components; unique number of the semantic part containing the mentioned concepts; unique number of the logical section containing the mentioned semantic part with concepts; a unique number of the logical construct containing the mentioned logical section and the mentioned semantic part with concepts.
  • the component association map consists of two parts - from the component combination map and the combination map semantically matching concepts.
  • a combination of components map can be used in a comprehensive legal analysis of some legal content, for example, an industry or thematic group of legal acts, or a separate legal act or a separate draft legal act.
  • the information registered in the Component Combination Map you can instantly get various “system-structural sections” of the legal acts (draft legal acts or group of legal acts) reflecting at least: the component structure of the rules (both at the level of the names used in the legal acts of the components and at the level types and values of components); the presence of separate combinations of components in the rules of the regulatory legal acts (also at three levels - name, type and value).
  • the first part of the communication map can be generated and presented in the form of a report.
  • This report can be generated according to the scheme and format, which will be set by the user in advance and will contain selective and system-selected information that clarifies the sections of the normative documents that are important for the user (draft legal acts, group of legal acts).
  • a map of combinations of semantically coinciding concepts can be used both in the legal analysis of certain legal content, for example, an industry or thematic group of legal acts, or a separate legal act or a separate draft legal act, and to search for related provisions in any legal content.
  • the simplest practical result of identifying relationships between prescriptions is to solve the problem of automatically identifying prescriptions establishing sanctions for violation of a rule specified in the original prescription and vice versa - to identify the relationship of the original prescription with the rule on the basis of the prescription containing sanctions for violation of this rule.
  • the related prescriptions identified in this way form a rule of law in which there are regulatory and protective parts.
  • FIG. 17 by way of example, but not limitation, a general flowchart of the inventive method 400 for finding relevant information in a transformed structured source data array containing at least logical constructs containing grammatically and spelling-correct semantic parts of logical sections of logical constructs is shown.
  • the claimed method 400 for searching relevant information is characterized by performing step 401 of generating a search query, which form a search query containing at least one search term and, if necessary, additionally assign at least one search criterion, which may be one from: search in the first and second semantic parts of the first logical sections of logical structures and the second semantic parts of the second logical sections of logical structures; search in the first semantic parts second logical sections of logical structures and second semantic parts of the second logical sections of logical structures; wherein the search query may be one of: at least one of: terminological search query; component search query; component terminological search query; a search array identification step 402, which identifies a data array in said transformed structured source data array corresponding to said search term; step 403 of the identification of semantic parts, which identify at least one grammatically and spelling correct semantic part of the logical sections of the logical structures of said transformed structured source data array containing at least said search term; a logical structure identification step 404, wherein at least one logical structure of said transformed structured
  • the claimed method 400 for searching for relevant information can be supplemented by a second search level, which consists in performing step 406 of selecting a logical structure, which selects at least one of these identified logical structures containing at least the aforementioned the grammatically and spelling-identified semantic part identified; step 407 of identifying related semantic parts, which identify at least associated with said grammatically and spelling correct semantic part grammatically and spelling correct semantic parts contained in other logical constructs of the converted structured data array, said identification being carried out on the basis of information from the map relations of components of grammatically and spelling-correct semantic parts of logical sections of logical constructs utechnisch; step 408 identification of logical structures, which identify at least one or more logical structures containing said identified associated grammatically and spelling correct semantic parts or identify the absence of logical constructions containing said identified associated grammatically and spelling-correct semantic parts; and step 409 of demonstrating the logical constructs, which demonstrate at least one or more logical constructs containing the identified identified gramm
  • the search array is a transformed structured source data array - i.e. normative legal acts in the form of logical constructions consisting of many data sub-arrays in which interdependencies exist both at the level of sub-arrays and between individual values of sub-arrays.
  • Sub-arrays of the transformed structured source data array are grammatically and spelling-correct semantic parts of logical sections of logical constructions.
  • Relevance of a search is the identification in the text of a normative legal act of logical constructions, which are normative prescriptions that correspond, at least, to the search query.
  • one of the possible search criteria that are intended to refine the search area can be additionally assigned.
  • a search criterion is assigned to the first and second semantic parts of the first logical sections of logical constructions and the second semantic parts of the second logical sections of logical constructions.
  • This search criterion allows you to search in arrays related to sections of situations. By assigning this search criterion, it will be possible to select the appropriate subarray (situations subarray) in which relevant information will be searched in accordance with the search query.
  • a search query using this search criterion can only be terminological.
  • a terminological query is a search query for matches with the term (concept) set in the query in the subarrays that match the search criteria.
  • a search criterion can be applied on the first semantic parts of the second logical sections of logical structures and the second semantic parts of the second logical sections of logical structures.
  • This search criterion allows you to search in arrays related to rule sections. By assigning this search criterion, it will be possible to select the appropriate subarray (subarray of rules) in which, in accordance with the search query, the relevant search will be performed information.
  • a search query can be simple and complex. Simple queries are a terminology query and a component query.
  • a component request is a request to search for matches, both by the presence of the components of a legal rule, and by their types, as well as by their values.
  • a component request can be three-level - search by the name of the component of the legal rule (first level of the component request), search by type of the component of the legal rule (second level of the component request), value of the type of component (third level of the component request)
  • the first two levels are search by the structure of the components of a legal rule.
  • the third level is a search by terms from the text of the LA, indexed in the corresponding subarrays as the value of the type of component of the legal rule.
  • a complex search query is a combination of two simple queries - component and terminological.
  • a component-terminological query is a component query, supplemented by a fourth search level - search by term or terms contained in the legal rules identified by the results of the component query.
  • performing a search query can be multi-stage.
  • the search criteria and the search query are applied to the entire base of legal entities.
  • the application of search criteria and search queries to the search results obtained by performing the first search stage can be carried out.
  • the sub-arrays of the transformed structured source data array corresponding to the search criterion match the indices and values with the set values of the search query.
  • the result of the search query is the identification of a list of individual subarray values that correspond to the values of the search query.
  • the values of subarrays identified as a result of the analysis of the transformed structured source array in accordance with the search criteria and the search query are associated with those semantic parts of the logical sections of the logical constructions in which they are registered.
  • those logical constructions are identified that contain semantic parts in which the values corresponding to the search query in subarrays matching the search criteria are identified.
  • the identified logical constructions - regulatory requirements will be demonstrated.
  • the format of logical constructions corresponds to the format of elements of the transformed structured source data array and contains legal rule and conditionality of the rule (conditionally - “situation”).
  • the legal rule consists of the semantic core of the legal rule and the refinement of the concepts of the semantic core of the rule
  • the conditionality of the rule consists of the semantic core of the situation, the refinement of the concepts of the semantic core of the situation, the dependencies of the concepts of the semantic core of the situation and the dependencies of the concepts of the semantic core of the rule.
  • search queries can be coordinated by the logical connection “and” (conjunction) or the logical connection or ”(disjunction).
  • conjunctive logical communication allows you to identify regulatory requirements that are fully consistent with all requests, i.e. regulatory requirements, which contain in the relevant search criteria parts of the text of the regulatory requirement, both the meaning of the first request and the values of the others associated with the first request.
  • disjunctive logical connection allows you to identify regulatory requirements that are fully consistent only with individual search queries, i.e. regulatory requirements, which contain in the relevant search criteria parts of the text of the regulatory requirement either the meaning of the first query or the meaning of other queries.
  • the search method may be supplemented by a second level of searching for relevant information.
  • the minimal task of identifying a logical rule of law is the task of identifying a normative prescription that establishes sanctions for violating a legal rule specified in the original logical construction (in the original normative prescription) or vice versa - identifying the normative prescription on the basis of the original logical construction containing sanctions.
  • a search query there may be either component values are used, or concepts (terms) that semantically coincide with the component values. It depends on what task is set when identifying the logical rule of law and this determines which arrays will be searched.
  • the search query is the values of the components of the rule, and the search array is the arrays of all semantic parts related to the conditions of the rule.
  • the search query When searching for the legal rule itself by the known, previously identified, sanctions, the search query are terms that semantically coincide with the values of the rule components, and the search array is the arrays of all semantic parts related to the legal rule.
  • logical constructs are identified that contain semantic parts that contain terms or components that correspond to the second level search query and which are associated with the original logical construct within the framework of the logical rule of law.
  • external relations of components are used, which make up the second part of the map of component relations (a map of combinations of semantically identical concepts), based on the presence of semantic relations between components and individual words, and phrases in semantic parts that do not contain components.
  • the semantic connection is understood as the presence between the meanings of the components and individual words, and phrases of semantic (semantic) coincidence, i.e. coincidence of the meaning of individual components and individual words within the meaning.
  • semantic semantic
  • semantically coinciding concepts can be not only completely matching or one-root words (terms), but also their synonyms and associations identified in special connected dictionaries and reference books. All concepts that semantically coincide with the values of the components identified in this way are recorded in a special table - a map of combinations of semantically identical concepts.
  • a map of combinations of semantically coinciding concepts establishes relationships between individual semantic parts containing components and semantic parts that do not contain components. This. the relationship is based on identified relationships between components and related concepts. As a result, related requirements are identified.
  • a map of combinations of semantically matching concepts contains the following information for each pair of related prescriptions: name of components; types of components to be combined; the value of the combined components; unique number of the semantic part containing the components; unique number of the logical section containing the mentioned semantic part; a unique logical construction number containing the logical partition; the name of concepts that semantically coincide with the meanings of the mentioned components; unique number of the semantic part containing the mentioned concepts; unique number of the logical section containing the mentioned semantic part with concepts; a unique number of the logical construct containing the mentioned logical section and the mentioned semantic part with concepts.
  • a map of combinations of semantically coinciding concepts can be used both in the legal analysis of certain legal content, for example, an industry or thematic group of legal acts, or a separate legal regulation or a separate draft legal regulation, and to search for related regulations at the request of the user in any legal content.
  • the user is shown the results in the form of identified related requirements (individual rules of law).
  • FIG. 19 by way of example, but not limitation, an exemplary diagram of the inventive system 500 is illustrated, which in a preferred embodiment includes at least one or more devices 501 comprising at least one or more processors 5011, one or more input modules I / O 5012 and memory 5013.
  • the devices 501 mentioned may include, but are not limited to: a personal computer, a laptop computer, a tablet computer, a PDA, a smartphone, a thin client, and the like.
  • the memory (computer-readable storage medium) 5013 of the device 501 contains a program code, which upon execution causes said one or more processors 5011 of said device 501 and / or device 501 associated therewith to perform the steps described in Embodiments 1 through 12 of the present invention, in Embodiments 20 through 21, and in embodiments the implementation of the present invention from the twenty-ninth to thirty-fourth, and contains subject to conversion or converted one or more structured source data arrays, containing at least natural language text.
  • one or more structured source data arrays to be converted or transformed can be downloadable and stored, in particular, in the database 02 of the structured data array transformation system.
  • a computer-readable storage medium may include random access memory (RAM); read-only memory device (ROM); Electrically Erasable Programmable Read-Only Memory (EEPROM); flash memory or other memory technologies; CDROM, digital versatile disc (DVD) or other optical or holographic storage media; magnetic cassettes, magnetic tape, magnetic disk storage device or other magnetic memory devices, wave carriers or other storage medium that can be used to encode the required information, and which can be accessed through the described device.
  • the memory includes a storage medium based on a computer storage device in the form of volatile or non-volatile memory, or a combination thereof. Exemplary hardware devices include solid state memory, hard disk drives, optical disk drives, etc.
  • An example environment is stored in the memory in which, using computer instructions or codes stored in the device’s memory, the procedures for pre-converting a structured source data array, the procedures for converting a structured source data array, forming a map of the components of the converted structured source data array, and the procedure can be performed search for relevant information in transformed structured source data arrays.
  • the device contains one or more processors 5011, which are designed to execute computer instructions or codes stored in the device memory in order to ensure that the above procedures are performed.
  • Computer commands or codes stored in memory are designed to perform preliminary transformations of a structured source data array, transformations structured source data array, forming a map of the components of the converted structured source data arrays, searching for relevant information in the converted structured source data arrays.
  • the I / O 5012 modules of device 501 are, but are not limited to, typical and prior art device controls: a mouse, keyboard, joystick, touchpad, trackball, electronic pen, stylus, touch screen, and the like. Also, I / O 5012 modules are, but are not limited to, typical and known from the prior art means of displaying information: a display, a monitor, a projector, a printer, a plotter, and the like.
  • the system 500 may also include a database (DB) 502.
  • the database 502 may include, but is not limited to: a hierarchical database, a network database, a relational database, an object database, an object-oriented database, an object-relational database, a spatial database, a combination of these two or more databases etc.
  • the database 502 stores data in memory, which may be, but not limited to: read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory, CDROM, digital versatile disk (DVD), or other optical or holographic storage media; magnetic cassettes, magnetic tape, magnetic disk storage device or other magnetic memory devices, wave carriers or other storage medium that can be used to store the required information, and which can be accessed by means of the device 501 converting a structured source data array and server 503.
  • ROM read-only memory
  • EEPROM electrically erasable programmable read-only memory
  • flash memory CDROM
  • DVD digital versatile disk
  • magnetic cassettes magnetic tape
  • magnetic disk storage device or other magnetic memory devices wave carriers or other storage medium that can be used to store the required information, and which can be accessed by means of the device 501 converting a structured source data array and server 503.
  • the database 502 is used to store data representing at least computer instructions or codes stored in memory, designed to perform pre Structured the transformations of the original data array transformation structured original data array components forming bonds card converted structured initial array data search the relevant information in converted structured source data arrays; to be converted or transformed one or more structured source data arrays containing at least natural language text that can be loaded into memory 5013 of device 501; and other data necessary for the functioning of the system.
  • An exemplary system 500 further comprises a server computing device (server) 503 that stores and facilitates manipulation of computer instructions or codes previously described herein, which, accordingly, are not further are described.
  • Server 503 may be a personal computer, a laptop computer, a tablet computer, a handheld computer, a smartphone, a database machine, and the like.
  • Server 503 provides data exchange control in the structured data source conversion system 500 and also provides data processing provided that one or more of the structured data array conversion devices 501 is connected to it or when the structured data array conversion device 501 is a thin client. In this case, all the computing power necessary to ensure the conversion of the structured data array is located on the server 503.
  • the system 500 also contains one or more data networks 504.
  • Data network 504 may include, but is not limited to, one or more local area networks (LANs) and / or wide area networks (WANs), or may be an information and telecommunications network Internet, or an Intranet, or a virtual private network (VPN) , or a combination thereof, and the like.
  • LANs local area networks
  • WANs wide area networks
  • VPN virtual private network
  • the server 503 also has the ability to provide a virtual computing environment (Virtual Machine) to facilitate interaction between the device 501 converting the structured data array and the database 502.
  • the network 504 is used to provide interaction between the device 501, the database 502, and the server 503 of the system 500.

Landscapes

  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Machine Translation (AREA)

Abstract

Le groupe d'invention concerne des solutions dans le domaine de traitement de collections de données, notamment des solutions dans le domaine du traitement de collections de données structurées contenant un texte en langage naturel et peut s'utiliser pour la transformation préliminaire nécessaire de la collection de données structurée, la formation d'une carte de liens des composants des parties de structures logiques de la collections de données structurée transformée et pour la recherche d'informations d'intérêt dans les collections de données structurées en utilisant les données provenant de la carte de liens des parties de structures logiques d'une collection de données structurée. L'invention est basée sur la mise en place d'un procédé de transformation préliminaire d'une collection de données structuré contenant un texte dans un langage structuré. Après avoir effectué une transformation préalable il est possible d'assurer la mise en place d'un procédé de formation d'une carte de liens des composants correctes du point de vue de la grammaire et de l'orthographe de parties logique des structures logiques d'une collection de données initiale structurée. Après avoir formé une carte de liens des composants de parties correctes du point de vue de la grammaire et de l'orthographe de parties logique des structures logiques d'une collection de données initiale structurée qui a été transformée, il est possible de réaliser une recherche d'informations désirées à l'intérieur de cette collection.
PCT/RU2015/000392 2014-06-27 2015-06-25 Procédé de transformation préliminaire d'une collection initiale de données, procédé de formation d'une carte de liens des composants de parties de structures logiques d'une collection de données initiale structurée qui a été transformée, procédé de recherche dans une collection de données initiale structurée qui a été transformée utilisant la carte de liens des composants et systèmes et dispositfs pour la mise en oeuvre de ces procédés WO2015199581A2 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EA201700031A EA201700031A1 (ru) 2014-06-27 2015-06-25 Способ предварительного преобразования исходного массива данных, способ формирования карты связей компонентов частей логических конструкций преобразованного структурированного исходного массива данных, способы поиска в преобразованном массиве данных с использованием карты связей компонентов и системы и устройства для реализации этих способов

Applications Claiming Priority (8)

Application Number Priority Date Filing Date Title
RU2014126199 2014-06-27
RU2014126198/08A RU2571406C1 (ru) 2014-06-27 2014-06-27 Способ двухуровневого поиска информации в предварительно преобразованном структурированном массиве данных
RU2014126198 2014-06-27
RU2014126195/08A RU2572367C1 (ru) 2014-06-27 2014-06-27 Способ поиска информации в предварительно преобразованном структурированном массиве данных
RU2014126197 2014-06-27
RU2014126199/08A RU2571407C1 (ru) 2014-06-27 2014-06-27 Способ формирования карты связей компонентов преобразованного структурированного массива данных
RU2014126197/08A RU2571405C1 (ru) 2014-06-27 2014-06-27 Способ предварительного преобразования структурированного массива данных
RU2014126195 2014-06-27

Publications (2)

Publication Number Publication Date
WO2015199581A2 true WO2015199581A2 (fr) 2015-12-30
WO2015199581A3 WO2015199581A3 (fr) 2016-02-25

Family

ID=54938906

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/RU2015/000392 WO2015199581A2 (fr) 2014-06-27 2015-06-25 Procédé de transformation préliminaire d'une collection initiale de données, procédé de formation d'une carte de liens des composants de parties de structures logiques d'une collection de données initiale structurée qui a été transformée, procédé de recherche dans une collection de données initiale structurée qui a été transformée utilisant la carte de liens des composants et systèmes et dispositfs pour la mise en oeuvre de ces procédés

Country Status (2)

Country Link
EA (1) EA201700031A1 (fr)
WO (1) WO2015199581A2 (fr)

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2392660C2 (ru) * 2008-04-15 2010-06-20 Государственное образовательное учреждение высшего профессионального образования "Мордовский государственный университет им. Н.П. Огарева" Способ поиска информации в массиве текстов
RU2399959C2 (ru) * 2008-10-29 2010-09-20 Закрытое акционерное общество "Авикомп Сервисез" Способ автоматизированной обработки текста на естественном языке путем его семантической индексации, способ автоматизированной обработки коллекции текстов на естественном языке путем их семантической индексации и машиночитаемые носители
US8335754B2 (en) * 2009-03-06 2012-12-18 Tagged, Inc. Representing a document using a semantic structure

Also Published As

Publication number Publication date
EA201700031A1 (ru) 2017-05-31
WO2015199581A3 (fr) 2016-02-25

Similar Documents

Publication Publication Date Title
US8341167B1 (en) Context based interactive search
US9633010B2 (en) Converting data into natural language form
JP2022547750A (ja) クロスドキュメントインテリジェントオーサリングおよび処理アシスタント
RU2564629C1 (ru) Способ кластеризации результатов поиска в зависимости от семантики
US9678949B2 (en) Vital text analytics system for the enhancement of requirements engineering documents and other documents
US10585893B2 (en) Data processing
US10339924B2 (en) Processing speech to text queries by optimizing conversion of speech queries to text
US8892575B2 (en) Dictionary entry name generator
US20180144065A1 (en) Method for Generating Visual Representations of Data Based on Controlled Natural Language Queries and System Thereof
Rockoff The language of SQL
US20070255694A1 (en) Document-drafting system using document components
US10706045B1 (en) Natural language querying of a data lake using contextualized knowledge bases
US20210182325A1 (en) Visual Presentation of Search Results
WO2015147706A9 (fr) Procédé de transformation d'un ensemble de données structuré
Kudashev Quality Assurance in Terminology Management
Khoo et al. Augmenting Dublin core digital library metadata with Dewey decimal classification
RU2571405C1 (ru) Способ предварительного преобразования структурированного массива данных
US10943073B2 (en) Dynamic candidate expectation prediction
US10467273B2 (en) Apparatus and method for displaying search results using cognitive pattern recognition in locating documents and information within
Melo et al. A strategy for archives metadata representation on CIDOC-CRM and knowledge discovery
US10089389B1 (en) Translation of unstructured text into semantic models
RU2571407C1 (ru) Способ формирования карты связей компонентов преобразованного структурированного массива данных
RU2572367C1 (ru) Способ поиска информации в предварительно преобразованном структурированном массиве данных
Gárdos et al. Identification of social scientifically relevant topics in an interview repository: a natural language processing experiment
RU2571406C1 (ru) Способ двухуровневого поиска информации в предварительно преобразованном структурированном массиве данных

Legal Events

Date Code Title Description
NENP Non-entry into the national phase in:

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 201700031

Country of ref document: EA

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 12.05.2017)

122 Ep: pct application non-entry in european phase

Ref document number: 15811548

Country of ref document: EP

Kind code of ref document: A2