Suitability of Machine Learning Methods for Prediction of Popularity on Social Media in Comparison of Different Data Sets

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62690094%3A18450%2F21%3A50018796" target="_blank" >RIV/62690094:18450/21:50018796 - isvavai.cz</a>

Výsledek na webu

<a href="https://uni.uhk.cz/hed/site/assets/files/1077/proceedings_2021_1-1.pdf" target="_blank" >https://uni.uhk.cz/hed/site/assets/files/1077/proceedings_2021_1-1.pdf</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.36689/uhk/hed/2021-01-025" target="_blank" >10.36689/uhk/hed/2021-01-025</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Suitability of Machine Learning Methods for Prediction of Popularity on Social Media in Comparison of Different Data Sets

Popis výsledku v původním jazyce

The growing popularity of various social media and the use of neural networks in recent years has brought predicting opportunities in various sectors. Social networks, in conjunction with neural networks, are often used in healthcare (prediction of whether a disease occurs or symptoms return) or business (what will be the profit or error rate of the products) and are playing an important economic and marketing change in the 21st century. The aim of the presented project is to contribute to a deeper understanding of which machine learning method to use for which data set in order to ensure the best possible prediction success across different industries. A total of 10 methods were used and the success of a total of 13 data files was tested with the help of specialized software for machine learning and the Python programming language. Of all the methods tested, the highest success rate on most datasets was with the Random Forrest method. The success rate ranged from 58.38% to 98.65%. Out of the total number of 13 datasets, the Random Forrest method was 5 times the best in accuracy.

Název v anglickém jazyce

Suitability of Machine Learning Methods for Prediction of Popularity on Social Media in Comparison of Different Data Sets

Popis výsledku anglicky

The growing popularity of various social media and the use of neural networks in recent years has brought predicting opportunities in various sectors. Social networks, in conjunction with neural networks, are often used in healthcare (prediction of whether a disease occurs or symptoms return) or business (what will be the profit or error rate of the products) and are playing an important economic and marketing change in the 21st century. The aim of the presented project is to contribute to a deeper understanding of which machine learning method to use for which data set in order to ensure the best possible prediction success across different industries. A total of 10 methods were used and the success of a total of 13 data files was tested with the help of specialized software for machine learning and the Python programming language. Of all the methods tested, the highest success rate on most datasets was with the Random Forrest method. The success rate ranged from 58.38% to 98.65%. Out of the total number of 13 datasets, the Random Forrest method was 5 times the best in accuracy.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10103 - Statistics and probability

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2021

Kód důvěrnosti údajů

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

Název statě ve sborníku

HRADEC ECONOMIC DAYS, VOL 11(1)

ISBN

978-80-7435-822-7

ISSN

2464-6059

e-ISSN

2464-6067

Počet stran výsledku

9

Strana od-do

251-259

Název nakladatele

Univerzita Hradec Králové

Místo vydání

Hradec Králové

Místo konání akce

Hradec Králové

Datum konání akce

25. 3. 2021

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000670596900025