Κατεύθυνση Οικονομικά, Διοικητικά και Πληροφοριακά Συστήματα Επιχειρήσεων
Library of the Faculty of Economics and of the Faculty of Business Administration

2018-11-21

2018

Zinonos Ioanna

Βασίλειος Λαζάρου, Διδάσκων, Τμήμα Οικονομικών Επιστημών, ΕΚΠΑ

Text mining: social network analysis for predicting consumer's trends within the IT industry

English
Greek

Text mining: social network analysis for predicting consumer’s trends within the IT industry

With the advancement of technology, more and more data is available in digital form. Among which, most of the data (approx. 85%) is in unstructured textual form. Text, so it has become essential to develop better techniques and algorithms to extract useful and interesting information from this large amount of textual data. Hence, the areas of text mining and information extraction have become quite popular areas of research, to extract interesting and useful information. Also sentiment analysis is widely used as part of the text mining technique to capture the emotions of people and convert them into meaningful conclusions and patterns helping businesses to take important precautions towards their strategic management. This paper focuses on the concept, process and applications of Text Mining via Social Network analysis for analyzing and predicting consumer’s trends within the IT sector. Algorithms and software tools will be used to extract data from social media applications such as twitter and with the right analysis such as sentiment analysis a statistical report and patterns will be created in order to draw conclusions on about which products consumers prefer and what their feelings are with the advancement of these products, also what else they would like to see in future concerning IT accessories and software. More specific, the collection of data from two social media platforms which is Twitter and Facebook with the use of R language is taken as a first step. These data that includes comments, hashtags, reactions and likes, are going under a sentiment analysis with the use of R programming language again so that the percentage of the opinions of the consumers for each brand and for every product of the brand will be calculated. The opinions are falling into three categories: negative, positive and neutral. Finally the results are transformed into statistical graphs in order to interpreter the results and drawn conclusions. To state which brand consumers prefer the most and which product are consumers more into it. Also for each product the variables that lead people to finalize a purchase and the sales volume for each brand are presented so that comparison tables can be drawn. After the analysis is completed and the conclusions about the effectiveness of text mining and sentiment analysis for marketing purposes some future suggestions on how the results of this master’s thesis can be evaluated and what can be integrated in the text mining and opinion mining field in future to evaluate the field as well will be presented too. Overall this thesis focuses more to present the effectiveness of text mining and sentiment analysis on social media for marketing purposes.

Technology - Computer science

Text Mining Algorithms, Data Mining, Information Retrieval, Information Extraction, social media, analytics, patterns, statistics, IT, software, consumers, trends

No

0

Yes

17

50

File access is restricted only to the intranet of UoA.

https://pergamos.lib.uoa.gr/uoa/dl/object/2819939