Sentiment analysis in Croatian tweets about earthquakes

Analiza sentimenta područje je koje je sve popularnije među istraživačima koji se bave obradom prirodnoga jezika. Ovo područje objedinjuje psihologiju, sociologiju, antropologiju, lingvistiku i računalnu tehnologiju te zahtijeva interdisciplinarni pristup. Zainteresiranost umjetnom inteligencijom i dubokim učenjem koje se može uočiti u svim granama računarstva nije zaobišla ni područje analize sentimenta. Međutim, alati izrađeni za analizu sentimenta u hrvatskome jeziku većinom su izrađeni koristeći tradicionalne i danas već zastarjele metode. U ovome se radu prikazuje postupak izrade alata za analizu sentimenta koji se temelji na suvremenijim metodama, na neuronskim mrežama. U prvome dijelu rada opisana su teorijska polazišta analize sentimenta i strojnoga učenja. Potom se opisuje izrada specijaliziranoga anotiranoga Twitter korpusa na temu potresa. Korpus je anotiralo dvoje autora te kao takav pruža dobar uvid u problematiku same jezične građe. Najčešće poteškoće koje se javljaju pri izradi ovakvoga jezičnoga resursa opisuju se u kratkoj diskusiji. U zadnjem se dijelu rada opisuje izrada alata za analizu sentimenta temeljen na rekurentnim neuronskim mrežama i treniran na već spomenutom korpusu. Rezultati se alata kontekstualiziraju s jedne strane u odnosu na slične radove, a s druge strane u odnosu na međuanotatorsku pouzdanost koja se računala za Twitter korpusa.Sentiment analysis is becoming more and more popular in the field of natural language processing. This specific field of study unites researchers with background in psychology, sociology, anthropology, linguistics, and computer sciences, and as such everyone interested in it has to have an interdisciplinary approach to the subject. The growing interest towards artificial intelligence and deep learning seen in every aspect of computer sciences did not bypass sentiment analysis either. However, tools developed to detect sentiment for Croatian language are mostly developed with traditional and outdated methods. In this paper we present the process of the development of an up-to-date sentiment analysis tool based on neural networks. The first part describes the theoretical framework of both sentiment analysis and deep learning, followed by the description of the making a specialized Twitter corpus about earthquakes. The corpus is annotated by two annotators and as such gives a unique perspective regarding the unusuality of the language used on Twitter. The most common difficulties in the process are briefly described as well. The final chapter of the paper describes the making of a sentiment analysis tool built with recurrent neural networks and trained on the mentioned corpus. The results of the sentiment analysis tool are contextualized both by comparing them to similar tools presented in other papers and the inter- annotator reliability calculated for Twitter corpus