Text classification based on topic modeling and chi-square

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27240%2F20%3A10246965" target="_blank" >RIV/61989100:27240/20:10246965 - isvavai.cz</a>

Result on the web

<a href="https://link.springer.com/chapter/10.1007%2F978-981-15-3308-2_56" target="_blank" >https://link.springer.com/chapter/10.1007%2F978-981-15-3308-2_56</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-981-15-3308-2_56" target="_blank" >10.1007/978-981-15-3308-2_56</a>

Result language

angličtina

Original language name

Text classification based on topic modeling and chi-square

Original language description

This paper compares two topic modeling algorithms - Latent Dirichlet Allocation (LDA), Latent Semantic Index (LSI), and a feature selection algorithm chi-square to extract news feature words. After feature extraction, the three classifiers (Logistics Regression, Naive Bayes and SVM) are compared in news classification. Based on the test results, combined LSI and Logistics Regression gives the highest result compared to the other algorithms, with precision of 96% and recall of 95%.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

S - Specificky vyzkum na vysokych skolach

Publication year

2020

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Advances in Intelligent Systems and Computing. Volume 1107

ISBN

978-981-15-3307-5

ISSN

2194-5357

e-ISSN

2194-5365

Number of pages

8

Pages from-to

513-520

Publisher name

Springer

Place of publication

Singapur

Event location

Čching-tao

Event date

Nov 1, 2019

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—