Εξόρυξη γνώσης από κείμενα του κοινωνικού ιστού
Date Issued
February 9, 2023
Type
Πτυχιακή Εργασία
Abstract
The technological conditions of the modern age and the digitization of every aspect of human life have resulted in the production of huge amounts of data, on a daily basis, by millions of people. These data contain hidden knowledge, which can be exploited by many different fields. For this reason, data mining techniques have been developed, combining different industries and technologies in order to find the hidden and potentially useful information contained in this data. Data mining is a process that consists of: 1) Requirements analysis, 2) data collection, 3) data selection, 4) data pre-processing, 5) data transformation, 6) data mining and 7) interpretation of results. Data mining can be done with many techniques, such as description, prediction, estimation, classification, clustering, regression, association rule mining, visualization and outlier detection.
The present study focuses on text mining from the social web, which is a rich source of information from texts produced by users. There are many applications for data mining on social media, such as sentiment analysis, competitive intelligence, cybersecurity, customer engagement, topic area recognition, and epidemic intelligence. For each of these applications, different tasks are presented in which they have been implemented.
Subjects