Open AccessConcept of weak-supervision approach use in multi-subject-domain information systems development.
Author(s) -
Maxim G. Shishaev,
Павел Ломов
Publication year - 2020
Publication title -
trudy kolʹskogo naučnogo centra ran
Language(s) - English
Resource type - Journals
ISSN - 2307-5252
DOI - 10.37614/2307-5252.2020.8.11.022
Subject(s) - subject (documents) , domain (mathematical analysis) , computer science , key (lock) , information system , development (topology) , data science , knowledge management , artificial intelligence , engineering , world wide web , mathematics , computer security , mathematical analysis , electrical engineering
The paper considers the concept of using a weak-supervision approach for the implementation of the components of the business logic of multi-subject-domain information systems. The specificity of multi-subject-domain information systems is characterized, key problems of heterogeneity and variability of system components and the external environment are identified. A brief description of the approach to machine learning based on inaccurate and incomplete data is given. The “model landscape” of a multi-subject-domain information systemusing weak-supervision is considered. The advantages and prospects of using the proposed approach are indicated.