Chart Classification Using Simplified VGG Model
| Autor: | Krešimir Nenadić, Filip Bajić, Josip Job |
|---|---|
| Přispěvatelé: | Snježana Rimac-Drlje, Drago Žagar, Irena Galić, Goran Martinović, Denis Vranješ, Marija Habijan |
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: |
Information retrieval
Contextual image classification Computer science business.industry 05 social sciences 020207 software engineering 02 engineering and technology Impaired Vision Visualization Search engine Data visualization Chart 0202 electrical engineering electronic engineering information engineering Key (cryptography) Visualization Chart Image Classification Convolutional Neural Networks Chart Recognition 0501 psychology and cognitive sciences The Internet business 050107 human factors |
| Zdroj: | IWSSIP |
| Popis: | Data visualization is created due to the need to show vast amount of data in a more transparent way. Data visualization often contains key information that is not listed anywhere in the text and allows the reader to find out important information and longer-term memory. On the other hand, Internet search engines have a problem with filtering data visualization and associating visualization and query that the user has entered. With the use of data visualization, all blind people and people with impaired vision are left off. This paper uses machine learning for classifying charts into 10 categories. The total accuracy achieved across all categories is 81.67%. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|