Константин Петров

Mr. Константин Петров

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
9 May 2021 in Remote Sensing
Verified
Journal Article
Open access
PDF file
VGValeri Gitis
Valeri Gitis
ADAlexander Derendyaev
Alexander Derendyaev
КПКонстантин Петров
Константин Петров
The results of earthquake prediction largely depend on the quality of data and the methods of their joint processing. At present, for a number of regions, it is possible, in addition to data from earthquake catalogs, to use space geodesy data obtained with the help of GPS. The purpose of our study is to evaluate the efficiency of using the time series of displacements of the Earth’s surface according to GPS data for the systematic prediction of earthquakes. The criterion of efficiency is the probability of successful prediction of an earthquake with a limited size of the alarm zone. We use a machine learning method, namely the method of the minimum area of alarm, to predict earthquakes with a magnitude greater than 6.0 and a hypocenter depth of up to 60 km, which occurred from 2016 to 2020 in Japan, and earthquakes with a magnitude greater than 5.5. and a hypocenter depth of up to 60 km, which happened from 2013 to 2020 in California. For each region, we compare the following results: random forecast of earthquakes, forecast obtained with the field of spatial density of earthquake epicenters, forecast obtained with spatio-temporal fields based on GPS data, based on seismological data, and based on combined GPS data and seismological data. The results confirm the effectiveness of using GPS data for the systematic prediction of earthquakes.

24 June 2022 in Journal of Communications Technology and Electronics
Verified
Journal Article
Closed access
PDF file
VGValeri Gitis
Valeri Gitis
ADAlexander Derendyaev
Alexander Derendyaev
КПКонстантин Петров
Константин Петров
Systematic prediction of earthquakes is regularly performed with step Δt in a predetermined zone of analysis. At each step, the training sample of target earthquake epicenters is supplemented, and new data about the seismic process is downloaded. The data is processed, converted into grid-based fields, and the machine-learning program calculates a solution using all available data from the starting time of training to the moment t* of the forecast. The result is a map of the alarm zone in which the appearance of the epicenter of the target earthquake is expected in the interval (t*, t* + Δt]. At the next step, the learning interval is increased by Δt. Training is based on the method of the minimum area of alarm. The mathematical model of the method makes it possible to select the outliers in the grid-based fields that precede target earthquakes, which may be precursors of earthquakes, to compare these outliers to each other, and then to use similar outliers in the forecast. The model introduces restrictions to the class of prediction decision rules. This helps counterbalance the small number of strong earthquakes used for training. In our previous works, the forecast quality was assessed by the proportion of detected target events at an assigned average value of the alarm zone. This work generalizes the method of the minimum area of alarm to improve another characteristic of forecast quality: the probability that at least one target event occurs in the projected alarm zone. The difference between the methods is that at the moment t* of the forecast, two decisions are made: forecast in time and forecast in space. The first decision determines the possibility of the occurrence of an earthquake epicenter with a target magnitude in the prediction interval. If it is decided that the target event is likely to occur, then a map with an alarm zone is calculated. A target earthquake is predicted if its epicenter falls into the calculated alarm zone. Predictive modeling is carried out for the regions of Kamchatka and California. An earthquake catalog is the initial data for Kamchatka. For the region of California, the data from an earthquake catalog and GPS time series are used. The results have shown quite high estimates of the probability of detecting target events and very low estimates of the probability of forecasting a target event epicenter occurrence in the alarm zone. At the same time, the estimates of the probability of systematic prediction are much higher than those for random fields.

12 September 2021
Verified
Book Chapter
Closed access
PDF file
VGValeri Gitis
Valeri Gitis
ADAlexander Derendyaev
Alexander Derendyaev
КПКонстантин Петров
Константин Петров

1 December 2020 in Journal of Communications Technology and Electronics
Verified
Journal Article
Closed access
PDF file
VGValeri Gitis
Valeri Gitis
ADAlexander Derendyaev
Alexander Derendyaev
КПКонстантин Петров
Константин Петров
In the paper, a new approach to identifying zones with rare anomalous manifestations of geological processes is proposed. The approach is based on two one-class classification methods of machine learning: the method of minimum area of alarm and the method of preference. The algorithm of minimum area of alarm is nonparametric. It is trained on a sample of anomalous events and computes the field of anomalous zones. The knowledge obtained by this method is non-verbalized. The method of preference allows approximating the obtained solution by a rather simple logical rule that defines the anomalous region in terms of analyzed properties of the geological environment. Examples of this approach to finding areas of possible foci of strong earthquakes and to make a regional forecast of deposits are considered.

23 July 2022
Verified
Book Chapter
Closed access
PDF file
VGValeri Gitis
Valeri Gitis
ADAlexander Derendyaev
Alexander Derendyaev
КПКонстантин Петров
Константин Петров

23 April 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
VGValeri Gitis
Valeri Gitis
ADAlexander Derendyaev
Alexander Derendyaev
КПКонстантин Петров
Константин Петров
EYEugene Yurkov
Eugene Yurkov
SPSergey Pirogov
Sergey Pirogov
AKAndrei Kaprin
Andrei Kaprin
Prostate cancer is the second most frequent malignancy (after lung cancer). Preoperative staging of PCa is the basis for the selection of adequate treatment tactics. In particular, an urgent problem is the classification of indolent and aggressive forms of PCa in patients with the initial stages of the tumor process. To solve this problem, we propose to use a new binary classification machine-learning method. The proposed method of monotonic functions uses a model in which the disease’s form is determined by the severity of the patient’s condition. It is assumed that the patient’s condition is the easier, the less the deviation of the indicators from the normal values inherent in healthy people. This assumption means that the severity (form) of the disease can be represented by monotonic functions from the values of the deviation of the patient’s indicators beyond the normal range. The method is used to solve the problem of classifying patients with indolent and aggressive forms of prostate cancer according to pretreatment data. The learning algorithm is nonparametric. At the same time, it allows an explanation of the classification results in the form of a logical function. To do this, you should indicate to the algorithm either the threshold value of the probability of successful classification of patients with an indolent form of PCa, or the threshold value of the probability of misclassification of patients with an aggressive form of PCa disease. The examples of logical rules given in the article show that they are quite simple and can be easily interpreted in terms of preoperative indicators of the form of the disease.