This paper introduces a novel natural language processing (NLP) model as an original approach to sentiment analysis, with a focus on understanding emotional responses during major disasters or conflicts. The model was created specifically for Croatian and is based on unigrams, but it can be used with any language that supports the n-gram model and expanded to multiple word sequences. The presented model generates a sentiment score aligned with discrete and dimensional emotion models, reliability metrics, and individual word scores using affective datasets Extended ANEW and NRC WordEmotion Association Lexicon. The sentiment analysis model incorporates different methodologies, including lexicon-based, machine learning, and hybrid approaches. The process of preprocessing includes translation, lemmatization, and data refinement, utilized automated translation services as well as the CLARIN Knowledge Centre for South Slavic languages (CLASSLA) library, with a particular emphasis on diacritical mark correction and tokenization. The presented model was experimentally evaluated on three simultaneous major natural crises that recently affected Croatia. The study’s findings reveal a significant shift in emotional dimensions during the COVID-19 pandemic, particularly a decrease in valence, arousal, and dominance, which corresponded with the two-month recovery period. Furthermore, the 2020 Croatian earthquakes elicited a wide range of negative discrete emotions, including anger, fear, and sadness, with the recuperation period much longer than in the case of COVID-19. This study represents an advancement in sentiment analysis, particularly in linguistically specific contexts, and provides insights into the emotional landscape shaped by major societal events.
For a number of years, various games have been used as an educational tool at different academic levels, especially in primary education. However, only recently games that teach coding and algorithmic thinking or even broader, computational thinking, emerged. Initiatives like Hour of Code and similar online activities or block-based programming games popularized the field, while at the same time, projects like CSUnplugged showed that the “idea of programming” can be learned even without using the computer. In this paper, we present our experience so far in creating and implementing different unplugged activities that teach students of different age levels, from early primary school to the university students, the simple programming concepts and algorithms. As a part of Science, Technology, Engineering, Mathematics and Computer Science (STEM-C) outreach program named SUZA - From school to science and the academic community at University of Zagreb Faculty of Electrical Engineering and Computing, Croatia, we conducted a number of game-like activities based on graph paper programming, block-based programming and using the existing board games. Although this is a work in progress project, the participant reactions so far, collected through interviews and questionnaires, indicated that the conducted activities were well accepted by students and their teachers. We also received a number of useful feedback comments and proposals, such as to expand the activities to include the homework part, or to make them more physical and relocate them outdoors.
The paper investigates the relationship between written reviews and numerical ratings of vegan and vegetarian restaurants, aiming to develop a predictive model that accurately determines numerical ratings based on review content. The dataset was obtained by scraping reviews from November 2022 until January 2023 from the TripAdvisor website. The study applies multidimensional scaling and clustering using the KNN algorithm to visually represent the textual data. Sentiment analysis and rating predictions are conducted using neural networks, support vector machines (SVM), random forest, Naïve Bayes, and BERT models. Text vectorization is accomplished through term frequency-inverse document frequency (TF-IDF) and global vectors (GloVe). The analysis identified three main topics related to vegan and vegetarian restaurant experiences: (1) restaurant ambiance, (2) personal feelings towards the experience, and (3) the food itself. The study processed a total of 33,439 reviews, identifying key aspects of the dining experience and testing various machine learning methods for sentiment and rating predictions. Among the models tested, BERT outperformed the others, and TF-IDF proved slightly more effective than GloVe for word representation.
This paper introduces a novel approach to the creation and application of confusion matrices for error pattern discovery in spellchecking for the Croatian language. The experimental dataset has been derived from a corpus of mistyped words and user corrections collected since 2008 using the Croatian spellchecker available at ispravi.me. The important role of confusion matrices in enhancing the precision of spellcheckers, particularly within the diverse linguistic context of the Croatian language, is investigated. Common causes of spelling errors, emphasizing the challenges posed by diacritic usage, have been identified and analyzed. This research contributes to the advancement of spellchecking technologies and provides a more comprehensive understanding of linguistic details, particularly in languages with diacritic-rich orthographies, like Croatian. The presented user-data-driven approach demonstrates the potential for custom spellchecking solutions, especially considering the ever-changing dynamics of language use in digital communication.
This paper presents a unique and extensive dataset containing over 33 million entries with pairs in the form “spelling error → correction” from ispravi.me, the most popular Croatian online spellchecking service, collected since 2008. The dataset, compiled from the contribution of nearly 900,000 users, is a valuable resource for researchers and developers in the field of natural language processing (NLP), improving spellcheck accuracy, and language learning applications. The dataset may be used to accomplish several goals: (1) improving spellchecking accuracy by incorporating common user corrections and reducing false positives and negatives; (2) helping language learners identify common errors and learn correct spelling through targeted feedback; (3) analyzing data trends and patterns to uncover the most common spelling errors and their underlying causes; (4) identifying and evaluating factors that influence typing input; (5) improving NLP applications such as text recognition and machine translation. Tasks specific to the Croatian language include the creation of a letter-level confusion matrix and the refinement of word suggestions based on historical usage of the service. This comprehensive dataset provides researchers and practitioners with a wealth of information, opening the path for advancements in spellchecking, language learning, and NLP applications in the Croatian language.