Powered by OpenAIRE graph
Found an issue? Give us feedback
addClaim

This Research product is the result of merged Research products in OpenAIRE.

You have already added 0 works in your ORCID record related to the merged Research product.

Оценка платежеспособности клиентов банка с использованием методов машинного обучения

выпускная квалификационная работа магистра

Оценка платежеспособности клиентов банка с использованием методов машинного обучения

Abstract

Тема выпускной квалификационной работы: «Оценка платежеспособности клиентов банка с использованием методов машинного обучения». Данная работа посвящена исследованию подходов к оценке платежеспособности банковских клиентов, с использованием методов машинного обучения. Задачи, которые решались в ходе исследования: Обзор литературы, посвященной использованию машинного обучения в банковской сфере; Изучение и подготовка набора данных для последующего анализа; Построение моделей оценки платежеспособности с использованием методов машинного обучения. Работа проведена с помощью языка R и интегрированной среды разработки RStudio, где проводилась значительная часть исследований. В ней проведен предварительный анализ набора данных, значения атрибутов преобразованы к корректным типам данных, показана структура набора данных. Оценка платежеспособности проводилась с использованием таких методов классификации, как логистическая регрессия, метод опорных векторов, алгоритм случайного леса и алгоритм деревьев решений. В результате проделанной работы было применено четыре метода машинного обучения для решения поставленной задачи. Для каждой модели получены графики ROC-кривых, показывающие качество прогнозируемой модели. Дополнительно приведены метрики качества полученных моделей, на основе которых сделан выбор наилучшей модели.

Theme of final qualification work: "Assessment of the solvency of bank customers using machine learning methods". This work is devoted to the study of approaches to assessing the solvency of bank customers using machine learning methods. Tasks that were solved during the study: Review of the literature on the use of machine learning in the banking sector; Study and preparation of a data set for subsequent analysis; Construction of solvency assessment models using machine learning methods. The work was carried out using the R language and the R Studio integrated development environment, where a significant part of the research was carried out. It contains a preliminary analysis of the data set, attribute values are converted to the correct data type, and the structure of the data set is shown. Solvency assessment was carried out using classification methods such as logistic regression, the method of support vectors, the algorithm of a random forest and the algorithm of decision trees. As a result of the work done, four machine learning methods were applied to solve the task. For each model, graphs of ROC curves are obtained, showing the quality of the predicted model. Additionally, the quality metrics of the obtained models are given, on the basis of which the choice of the best model is made.

Keywords

классификация, Искусственный интеллект, classification, логистическая регрессия, logistic regression, scoring, случайный лес, скоринг, random forest

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Upload OA version
Are you the author of this publication? Upload your Open Access version to Zenodo!
It’s fast and easy, just two clicks!