GR Semicolon EN

Show simple item record

dc.contributor.author
Nissopoulou, Theopisti Xeni
en
dc.date.accessioned
2023-04-11T12:09:35Z
dc.date.available
2023-04-11T12:09:35Z
dc.date.issued
2023-04-11
dc.identifier.uri
https://repository.ihu.edu.gr//xmlui/handle/11544/30145
dc.rights
Default License
dc.subject
Web content
en
dc.subject
Classification analysis
el
dc.title
Web content classification analysis
en
heal.type
masterThesis
en_US
heal.dateAvailable
2023-01-01
heal.language
en
en_US
heal.access
free
en_US
heal.license
http://creativecommons.org/licenses/by-nc/4.0
en_US
heal.recordProvider
School of Science and Technology, MSc in Data Science
en_US
heal.publicationDate
2023-01-07
heal.abstract
This dissertation was written as a part of the MSc in Data Science at the International Hellenic University. Web analytics is a way for companies to learn more about the people who visit their websites. This information may include things like how people use the pages on a site, and how they respond to different types of content. These types of analytics can help to determine future decisions regarding the content and marketing which may help on how the company is perceived from the customers, as well as it can even improve their status and the profit. In order to gain valuable insights from a series of site visits and other related interaction data, it is essential to have accurate data. In the current work, Web Content Text Classification is to be done over the extracted content of a company’s portal by categorizing those into 19 brands. Natural Language Processing techniques and Machine Learning algorithms have been applied and described. After evaluating the results of the models, it was concluded that BERT, which is a powerful deep learning model, and in particular Bert Base Uncased, performed the best, making accurate predictions and having good performance overall.
el
heal.advisorName
Karapiperis, Dimitrios
el
heal.committeeMemberName
Diamantaras, Konstantinos
en
heal.academicPublisher
IHU
el
heal.academicPublisherID
ihu
en_US


This item appears in the following Collection(s)

Show simple item record

Related Items