Search academic texts

Information × Registration Number 0224U031670, 0119U002279 , R & D reports Title To develop distributed machine learning methods for genetic codes noise immunity analysis popup.stage_title Head Gupal Anatoliy M., Доктор фізико-математичних наук Registration Date 03-06-2024 Organization V.M. Glushkov Institute of Cybernetics of the National Academy of Sciences of Ukraine popup.description2  Distributed machine learning methods for Bayesian recognition procedures using the MapReduce programming system have been developed. Bayesian procedures are easily adapted to the parallel mode of calculation, minimize the exchange of information between network nodes and do not change the structure of training samples with the appearance of new data. Fast Bayesian methods of analysis and recognition of intracranial neoplasms and craniocerebral injuries have been developed based on three structures of input data (blood parameters), which have been converted into a digital format. Based on this approach, it was possible to determine the degree of malignancy of neoplasms for the first time. The immunity of the standard code with respect to the polarity of amino acids has been established. Optimal interference-resistant codes are built. With the help of genetic algorithms, various versions of the most interference-resistant codes with characteristics different from those of the standard code were built. Based on databases of genetic diseases using standard and optimal codes, it is shown that the immunity of the code has a significant effect on the violation of the polarity of amino acids in the case of point mutations of nucleotides. Product Description popup.authors Biletskyi Borys O. Byts Oleksii V. Boiko Valentyna V. Hrachova Tamara Ya. Tarasov Andrii L. popup.nrat_date 2024-06-03 Close