|
|
Analýza sentimentu textových recenzií vybraného segmentu produktov
Košťálová, Dorota
This bachelor thesis deals with the analysis of the sentiment of professional reviews of mobile phones. It focuses on reviews in the Czech language. The theoretical part of the work describes natural language processing, procedures and tools used in this area. The main part of the work deals with the selection of a suitable tool for sentiment analysis. In the experiments, three approaches to analysis of reviews were tried. Subsequently, a module was created that is able to determine the sentiment of the input dataset. In the last part of the work, possible improvements were suggested that would help to achieve better results.
|
|
|
Analýza sentimentu textových recenzií vybraných kategorií produktov
Mikula, Michal
With still increasing popularity of e-shops, social media and mobile devices, the amount of user generated text content is rapidly rising. Majority of this text is taking form of unstructured data. Until recently, evaluation of this data was very challenging, as it required great amount of time and manual human analysis. Because of that, the field of artificial inteligence, natural language processing (NLP), is becoming more and more relevant. Using machine learning methods in combination with elements from linguistics and computer science NLP creates a system capable of understanding, analyzing and extracting meaning from the text. Sentiment analysis, also known as opinion mining is an NLP technique, which can decide whether the emotional sentiment of the given data is positive, negative or neutral.
|
|
|
Detection of Intensity in Sentiment Analysis of Czech
Dargaj, Jakub ; Tamchyna, Aleš (advisor) ; Mareček, David (referee)
Sentiment analysis is concerned with automatic extraction of subjective information from text. The goal of this thesis is to predict the intensity of attitude in Czech texts. In order to solve this task, we prepared a dataset of movie reviews by users of Czech-Slovak Film Database. We compare several machine learning methods, focusing on feature extraction from text data. Using convolutional neural networks and corpus-dependent training of word embeddings, we surpassed basic models and achieved accuracy similar to the most recent results in this field. We also analyze the logistic regression model in order to compare the vocabulary used in reviews with different ratings.
|
|
|
Impact of European Central Bank and Federal Reserve System statements on cryptocurrency markets via sentiment analysis
Krejcar, Vilém ; Krištoufek, Ladislav (advisor) ; Čech, František (referee)
This study explores the impact of public statements from major central banks, specifically the FED and the ECB, on Bitcoin volatility from 2018 to 2021. Utilizing high-frequency data, we computed Bitcoin's volatility and extracted sentiment scores from the central banks' communications using two methods: the FinBERT language model and the state-of-the-art Generative AI GPT-4 model with tailored prompt. The GPT-4 model, capturing more nuanced senti- ment from text, was deemed superior. Our analysis involved comparing various models, with the HAR model emerging as the most e ective for this study. The research findings are particularly significant: negative sentiment from the ECB during the pandemic was associated with immediate and significant increases in Bitcoin volatility, indicating a market reaction of caution when faced with negative emission. These findings highlight the significant impact of central bank sentiment on Bitcoin volatility, confirming the initial hypothesis of this research. Additionally, the results provide a motivation to incorporate Genera- tive Artificial Intelligence into academic research as a tool for uncovering novel insights. JEL Classification C32, C55, C58, E58, G15 Keywords central banks, sentiment analysis, volatility, Bit- coin, GenAI, HAR, FED, ECB Title Impact of European...
|
|
|
Interviews 2.0 - Using AI for oral historians
Haubert, Marek ; Hlaváček, Jiří (advisor) ; Wohlmuth Markupová, Jana (referee)
This diploma thesis explores the integration of oral history with modern information technologies (IT), especially Artificial Intelligence (AI), aiming to investigate how these technologies can enrich the practice of oral historians and make the processing of oral historical interviews more efficient. It demonstrates, through practical examples, the possibilities of integrating AI and IT services at all stages of oral historical research, from interview preparation to realization, subsequent transcription, analysis, interpretation, and up to its security, archiving, and public publication. The thesis emphasizes practical demonstrations of technology use and research on available services and tools that can facilitate recording interviews, their transcription, sentiment analysis, or metadata creation.
|
|
|
Metasearch for Reviews on the Czech Web
Šmahel, Michal ; Doležal, Jan (referee) ; Smrž, Pavel (advisor)
The main purpose of this work is to create a metasearch engine for review articles with built-in sentiment analysis. In addition, a complex survey of main text extraction tools and web browser automation tools for web crawling has been carried out to achieve of the best possible results. The resulting metasearch engine provides a web interface for searching relevant review articles, thus saving time spent on manual searching. Thanks to multi-level transformer-based filtering, it can return 10—15 relevant review articles on frequently reviewed topics in about 4 minutes with no effort, just by clicking on a button.
|
|
|
Analysis of Product Reviews
Klocok, Andrej ; Doležal, Jan (referee) ; Smrž, Pavel (advisor)
Online store customers generate vast amounts of product and service information through reviews, which are an important source of feedback. This thesis deals with the creation of a system for the analysis of product and shop reviews in the czech language. It describes the current methods of sentiment analysis and builds on current solutions. The resulting system implements automatic data download and their indexing, subsequently sentiment analysis together with text summary in the form of clustering of similar sentences based on vector representation of the text. A graphical user interface in the form of a web page is also included. A review data set with a total of more than six million reviews was created during the semester along with an interface for easy data export.
|
|
|
Analysis of Social Media Content Discussing Czech Mobile Operators
Pavlů, Jan ; Otrusina, Lubomír (referee) ; Smrž, Pavel (advisor)
The main topic of this thesis is sentiment analysis of posts obtained from a social networks. The posts are about czech mobile network operators. The essential part of implemented system is also data visualization. The sentiment analysis is done using machine learning techniques. Downloaded posts are cleaned, lemmatized and transformed to feature vectors. Stochastic Gradient Descent algorithm is used for classification. Analyzed data are visualized in charts and as the list of posts. The system provides tools for text categorization. The accuracy, precision, recall and F1 score of sentiment analysis is about 75%. The accuracy of post categorization is high (about 80%), but precision, recall and F1 score are low (about 30%). This is the reason why post categorization isn't automatically done. The benefit of the system it that it automatically collects data from different sources, analysis them and displays them. It also provides tools for manual change of sentiment/categories which can lead to better system characteristics with some help of users.
|
|
|
Sentiment Analysis with Use of Data Mining
Sychra, Martin ; Burget, Radek (referee) ; Bartík, Vladimír (advisor)
The theme of the work is sentiment analysis, especially in terms of informatics (marginally from a linguistic point of view). The linguistic part discusses the term sentiment and language methods for its analysis, e.g. lemmatization, POS tagging, using the list of stopwords etc. More attention is paid to the structure of the sentiment analyzer which is based on some of the machine learning methods (support vector machines, Naive Bayes and maximum entropy classification). On the basis of the theoretical background, a functional analyzer is projected and implemented. The experiments are focused mainly on comparing the classification methods and on the benefits of using the individual preprocessing methods. The success rate of the constructed classifier reaches up to 84 % in the cross-validation.
|
|
|
Sentiment Analysis in Automotive Industry
Bezák, Adam ; Otrusina, Lubomír (referee) ; Smrž, Pavel (advisor)
The main theme of this thesis is to familiarize with the basic methods of sentiment analysis on social networks. Thesis’s theme is aimed on the automotive industry, although this prinicipal can be used in any different examined branch. The basis of the practical part is to obtain data from the social networks, analyze them and then index them into ElasticSearch database. Another goal of the thesis is to visualize these data by means of a web portal. Created web portal provides various statistics of the leading automobile brands, an overview of new trends or the aspect visualization of the individual cars.
|
guest ::
