On a Weakly Supervised Classification Problem

Vladimir Berikov, Alexander Litvinenko, Igor Pestunov, Yuriy Sinyavskiy

Результат исследования: Публикации в книгах, отчётах, сборниках, трудах конференцийстатья в сборнике материалов конференциинаучнаярецензирование

Аннотация

We consider a weakly supervised classification problem. It is a classification problem where the target variable can be unknown or uncertain for some subset of samples. This problem appears when the labeling is impossible, time-consuming, or expensive. Noisy measurements and lack of data may prevent accurate labeling. Our task is to build an optimal classification function. For this, we construct and minimize a specific objective function, which includes the fitting error on labeled data and a smoothness term. Next, we use covariance and radial basis functions to define the degree of similarity between points. The further process involves the repeated solution of an extensive linear system with the graph Laplacian operator. To speed up this solution process, we introduce low-rank approximation techniques. We call the resulting algorithm WSC-LR. Then we use the WSC-LR algorithm for analysis CT brain scans to recognize ischemic stroke disease. We also compare WSC-LR with other well-known machine learning algorithms.

Язык оригиналаанглийский
Название основной публикацииAnalysis of Images, Social Networks and Texts - 10th International Conference, AIST 2021, Revised Selected Papers
РедакторыEvgeny Burnaev, Sergei Ivanov, Alexander Panchenko, Dmitry I. Ignatov, Sergei O. Kuznetsov, Michael Khachay, Olessia Koltsova, Andrei Kutuzov, Natalia Loukachevitch, Amedeo Napoli, Panos M. Pardalos, Jari Saramäki, Andrey V. Savchenko, Evgenii Tsymbalov, Elena Tutubalina
ИздательSpringer Science and Business Media Deutschland GmbH
Страницы315-329
Число страниц15
ISBN (печатное издание)9783031164996
DOI
СостояниеОпубликовано - 2022
Событие10th International Conference on Analysis of Images, Social Networks and Texts, AIST 2021 - Tbilisi, Грузия
Продолжительность: 16 дек. 202118 дек. 2021

Серия публикаций

НазваниеLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Том13217 LNCS
ISSN (печатное издание)0302-9743
ISSN (электронное издание)1611-3349

Конференция

Конференция10th International Conference on Analysis of Images, Social Networks and Texts, AIST 2021
Страна/TерриторияГрузия
ГородTbilisi
Период16.12.202118.12.2021

Предметные области OECD FOS+WOS

  • 1.02 КОМПЬЮТЕРНЫЕ И ИНФОРМАЦИОННЫЕ НАУКИ

Fingerprint

Подробные сведения о темах исследования «On a Weakly Supervised Classification Problem». Вместе они формируют уникальный семантический отпечаток (fingerprint).

Цитировать