FIELD: computer engineering.
SUBSTANCE: technical result is achieved due to a text processing system for generating an ESG rating, comprising a web server, a data storage module, a computing module, a user terminal, a data bus, wherein the computing module consists of operating units connected to each other through inputs-outputs and connected in two modes: in the setting mode, the following are connected in series: a unit for forming topics, a unit for forming dictionaries, a unit for forming a training sample, a unit for forming a matrix of features, a unit for forming a reference sample and training a rating model, in the rating mode, the following are connected in series: a data preparation unit, a data preprocessing unit, a topic evaluation unit, a rating unit, a results output unit, at that, in the data preparation unit, a list of data sources and a list of text materials characterizing the assessed companies are formed, wherein in the data preprocessing unit, the texts are broken down into units of at least 2,000 characters, wherein smaller parts of the text are excluded from consideration.
EFFECT: high accuracy of processing texts for generating an ESG rating.
2 cl, 3 dwg, 7 tbl
Title | Year | Author | Number |
---|---|---|---|
METHOD AND SYSTEM FOR CLASSIFYING AND FILTERING PROHIBITED CONTENT IN A NETWORK | 2020 |
|
RU2738335C1 |
SYSTEM AND METHOD FOR DETERMINATION OF EVENT CLASSIFICATION RULE ON USER TERMINAL DEVICE | 2020 |
|
RU2772404C2 |
METHOD AND SYSTEM FOR ARRANGING DIALOGUE WITH USER IN USER-FRIENDLY CHANNEL | 2018 |
|
RU2688758C1 |
METHOD FOR ATTRIBUTION OF PARTIALLY STRUCTURED TEXTS FOR FORMATION OF NORMATIVE-REFERENCE INFORMATION | 2020 |
|
RU2750852C1 |
METHOD FOR AUTOMATIC ITERATIVE CLUSTERISATION OF ELECTRONIC DOCUMENTS ACCORDING TO SEMANTIC SIMILARITY, METHOD FOR SEARCH IN PLURALITY OF DOCUMENTS CLUSTERED ACCORDING TO SEMANTIC SIMILARITY AND COMPUTER-READABLE MEDIA | 2014 |
|
RU2556425C1 |
USE OF AUTOENCODERS FOR LEARNING TEXT CLASSIFIERS IN NATURAL LANGUAGE | 2017 |
|
RU2678716C1 |
AUTOMATIC DETERMINATION OF SET OF CATEGORIES FOR DOCUMENT CLASSIFICATION | 2018 |
|
RU2701995C2 |
METHOD AND SYSTEM FOR IDENTIFYING EXPLOITED VULNERABILITIES IN THE PROGRAM CODE | 2022 |
|
RU2790005C1 |
METHOD AND SYSTEM FOR CHECKING AN ELECTRONIC SET OF DOCUMENTS | 2019 |
|
RU2702967C1 |
METHOD AND SYSTEM FOR SUPPORTING MEDICAL DECISION MAKING USING MATHEMATICAL MODELS OF PRESENTING PATIENTS | 2017 |
|
RU2703679C2 |
Authors
Dates
2024-08-19—Published
2023-12-11—Filed