=Paper=
{{Paper
|id=Vol-3293/paper61
|storemode=property
|title=Natural Language Processing Tools for Performing Effective Text Mining Tasks in Greek Food & Beverage Sector - Abstract
|pdfUrl=https://ceur-ws.org/Vol-3293/paper61.pdf
|volume=Vol-3293
|authors=Anastasios Liapakis,Theodore Tsiligiridis,Constantine Yialouris,Constantina Costopoulou,Kyvele-Constantina Diareme,Pagona Gorou
|dblpUrl=https://dblp.org/rec/conf/haicta/LiapakisTYCDG22
}}
==Natural Language Processing Tools for Performing Effective Text Mining Tasks in Greek Food & Beverage Sector - Abstract==
<pdf width="1500px">https://ceur-ws.org/Vol-3293/paper61.pdf</pdf>
<pre>
Natural Language Processing Tools for Performing Effective
Text Mining Tasks in Greek Food & Beverage Sector - Abstract
Anastasios Liapakis 1,2,3, Theodore Tsiligiridis 2, Constantine Yialouris 2, Constantina
Costopoulou 2, Kyvele-Constantina Diareme 2,3 and Pagona Gorou 3
1
  National & Kapodistrian University of Athens, Dept. of Digital Industry Technologies, Psachna, Chaklis,
Greece
2
  Agricultural University of Athens, Dept. of Agricultural Economics & Rural Development, Informatics
Laboratory, Iera Odos 75, Athens, Greece
3
  New York College of Greece, Dept. of Informatics, Leoforos Vasilisis Amalias 38, Athens, Greece


                 Summary
                 Nowadays, more and more companies use the social media networking to attract more
                 customers. This modifies consumers’ attitudes and companies, or other stakeholders cannot
                 detect these modifications due to the big volume and the diversity of the produced information.
                 In the case of the Food and Beverage (F&B) sector, which is one of the most dynamic sectors
                 in Greece, the use of social media networks is very high for multinational and large companies.
                 Delivery or take away food or coffee is very common, with the vast majority of consumers to
                 order from aggregators’ platforms (online digital markets). Thus, a large amount of data
                 containing useful information concerning the consumers’ preferences is generated from these
                 online digital markets. The produced data (evaluations) which is generated rapidly can be large
                 and cannot be mined and analyzed in real-time due to the lack of resources. Greek and many
                 other European languages, show a low density of linguistic resources and knowledge bases,
                 making it difficult to perform text mining and natural language processing tasks. The situation
                 is getting even more difficult by the fact that Greek is a high-dimensional language with a lot
                 of complex grammatical and syntax rules. The purpose of this research is to propose some
                 Natural Language Processing tools for performing effective text mining tasks in the Greek
                 Language helping the stakeholders in extracting, analyzing, and inferring meaningful
                 information. The tools are tested in a dataset that contains 80,500 customers’ reviews written
                 in Greek Language and the findings will be practical and significant, as not enough attention
                 has been paid to Natural Language Processing techniques and tools used in combination with
                 non-English, like the modern Greek language.

                 Keywords 1
                 Analytics and Data Science, Social Networks, Natural Language Processing, Computational
                 Linguistics, Sentiment Analysis, Food & Beverage Sector, Greek Language




Proceedings of HAICTA 2022, September 22–25, 2022, Athens, Greece
EMAIL: anliapakis@dind.uoa.gr (A. 1); tsili@aua.gr (A. 2); yialouris@aua.gr (A. 3); tina@aua.gr (A. 4); kkdiareme@aua.gr (A. 5);
ngorou@nyc.gr (A. 6)
ORCID: 0000-0003-2183-4760 (A. 2); 0000-0001-5070-3693 (A. 3); 0000-0003-0151-5586 (A. 4); 0000-0002-3478-4697 (A. 5)
              ©️ 2022 Copyright for this paper by its authors.
              Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
              CEUR Workshop Proceedings (CEUR-WS.org)




                                                                                   329

</pre>