23.12.2019
One of the activities of the group of companies is the development of intelligent solutions for automating processes related to auditing and consulting activities, and the implementation of compliance procedures at enterprises.
Starting in December 2019, a team of specialists has been dedicated to developing fundamentally new technologies for parsing document meanings with one button, designed to reduce user labor costs from several hours/days to several seconds.
The solution uses proprietary machine learning algorithms for pattern recognition, image preprocessing, document classification through clustering, automated learning through taking a sample of a set of labeled documents with representative datasets, and machine learning through a continuous feedback loop during processing operations. documents.
An approach with pre-trained language models in architecture is used to extract information from documentsTransformer. This architecture was first introduced by Google, but in this project it was supplemented with layers, and the language model used was adapted to the specifics of the recognized documents.
The development is supervised by the scientific supervisor Irina Anatolyevna Arekhina, PhD in Economics with many years of experience in management, auditing, as well as research and practical work on setting up a budgeting system, managerial accounting, and financial management development, in the field of developing practical recommendations for management of regional and municipal finances, development of methodological and practical recommendations for credit institutions of all levels, practical experience in training and advising employees of financial services.
