Verslo ir ekonomikos institutas / Institute of Busines and Economics
Contactless Fall Detection Using mmWave Radar with Stacking-Based Ensemble LearningItem type:Publication, research article[2025][S5][N009,T007,S003][9] ;Lee, Young Sil; ; Jung, Sang-JoongThe Journal of Korea Institute of Convergence Signal Processing., 2025, p. 315-323With the transition to an aging society, increasing attention has been directed toward early detection of falls among elderly individuals. Vision-based fall detection systems suffer from privacy concerns and environmental sensitivity, while wearable sensor-based approaches impose a user burden. To address these limitations, this paper proposes a contactless fall detection method using millimeter-wave (mmWave) radar. Radar data collected in an indoor environment are used to design a stacking-based ensemble learning framework that combines traditional machine learning and deep learning models. The proposed method integrates predictions from multiple base classifiers through a meta-classifier to improve detection performance and robustness. Experimental results using real-world radar data demonstrate that the proposed ensemble model outperforms conventional single classifiers across accuracy, precision, recall, and F1-score. These findings indicate that radar-based sensing, combined with ensemble learning, provides a practical, privacy-preserving solution for indoor fall detection.
22 LLM-Based Learning for Enhancing Interaction Between Game Players and NPCsItem type:Publication, research article[2025][S5][N009,T007,S003][9] ;Huofeng, Li; ;Zhou, JianiYu, TaesooThe Journal of Korea Institute of Convergence Signal Processing., 2025, p. 191-199In today's games, traditional non-player character (NPC) behaviors and dialogues rely on preset scripts, making it difficult to dynamically respond to diverse player needs and limiting the immersion and interactive depth of the game. This study aims to explore the application of Large Language Models(LLM) in games to enhance the intelligent interaction of NPCs. In this study, the Llama model is fine-tuned to build a simulated game scene in Unity engine, and the fine-tuned model is embedded into an NPC system to verify its performance and ability to drive NPCs in real game interactions. In the experiments, the research team designed a series of typical game interaction scenarios, including item collection, task delivery, and complex semantic Q&A, through which the NPC's language comprehension and interaction capabilities were tested. The results demonstrate that LLM-driven NPCs enable game developers to build more flexible and intelligent interaction mechanisms. This approach opens new possibilities for intelligent NPC development.
2 Prediction of the Main Environmental and Energy Characteristics of Diesel Engines using an Artificial Neural Network for Pure Diesel Fuel, Pure HVO, and a Mixture of these Fuels in the Ratio 50/50 by VolumeItem type:Publication, research article[2024][S1][N009,T003,T001][14] ;Matijošius, Jonas ;Rimkus, AlfredasTransport problems = Problemy transportu., 2024, p. 69-82This research assesses the influence of different quantities of hydrotreated vegetable oil (HVO) in diesel fuel on the performance of the engine and the emissions it produces. The particular areas of interest are the level of smoke emitted and the brake thermal efficiency (BTE). A series of engine experiments were undertaken to quantify emissions and performance characteristics under various operating regimes using three distinct fuel blends: D100 (pure diesel), HVO50 (50% HVO mixture), and HVO100 (pure HVO). The results show a tendency to reduce soot emissions when the amount of HVO in the mixture reaches 50%, but in order to accurately determine the correlation between the amount of HVO and emissions, additional studies with various concentrations of HVO and diesel mixtures are necessary. Similarly, the results show a slight improvement in BTE stability with a 50% HVO blend, but more studies with different percentages of HVOdiesel blends are needed to reliably determine changes in BTE. This indicates a trend of change in engine performance with increasing HVO concentration in diesel fuel. In order to forecast emissions and performance indicators under different operating scenarios, we used artificial neural networks ANNs, which demonstrated excellent prediction accuracy, exhibiting robust linear correlations between the expected and real values for all fuel types. This research emphasizes the advantages of utilizing HVO in diesel engines for both environmental impact and performance. It also emphasizes the usefulness of ANNs in optimizing engine settings to improve efficiency and minimize emissions. The results endorse the further use of HVO as a viable substitute for conventional diesel, leading to less ecological consequences and enhanced engine efficiency.
9Scopus© Citations 3 Use of cookies after GDPR: A case study of top Lithuanian websitesItem type:Publication, [Slapukų naudojimas po BDAR: populiariausių Lietuvos interneto svetainių atvejo tyrimas]research article[2021][S1][S003,N009][24]; ;Driaunys, KęstutisTransformations in business and economics. Kaunas : Kaunas Faculty of Humanities. Vilnius University, 2021, vol. 20, iss. 3., p. 93-116With the implementation of the European Union’s General Data Protection Regulation (GDPR) on 25 May 2018, a new regulatory regime for business in Europe and beyond has begun. Many companies had to adjust their data processing processes to meet GDPR compliance requirements. GDPR stipulates that data controllers must have a legal basis for the collection and processing of personal data. One of the most common ways for personal data to be collected and shared online is through website cookies. A cookie is a small text file that is downloaded onto ‘terminal equipment’ when the user accesses a website. It allows the website to recognise that user’s device and store some information about the user’s preferences or past actions. Cookies, insofar as they are used to identify users, qualify as personal data and are therefore subject to the GDPR. The investigative problem of this study is expressed through the following question: Which requirements are applied to cookie usage now that the GDPR has come into effect, and what impact has this had on Lithuania’s top popular websites? The main purpose of this article is to assess the tendencies as regards the usage of the cookies in accordance with GDPR requirements on the most popular Lithuanian websites. The methodological framework used in the investigation is based on scientific literature analysis and the qualitative research method. The work carried out theoretical narrative, systematic, comparative analysis, and generalisation. A qualitative case study was carried out examining the 100 most popular Lithuanian websites. The obtained empirical findings demonstrate that Lithuanian websites use cookies in breach of GDPR requirements and that complex violations occur. In Lithuania, unlike the other EU Member States, violations pertaining to cookie management continue to remain unnoticed, organisations do not abide by GDPR requirements en masse, whereas individual users and the State Data Protection Inspection are merely passive observers. The conclusions and recommendations of the research have practical value, which will help organisations, users and supervisory authorities to identify the violations connected to cookie usage and to take the relevant measures in order to ensure that data are being processed lawfully and in keeping with GDPR requirements.
35Scopus© Citations 2 Development of digital contents: privacy policy model under the general data protection regulation and user-friendly interfaceItem type:Publication, research article[2020][S1][S003,S001,N009][22]; ;Driaunys, Kęstutis; Transformations in business and economics. Kaunas : Vilnius University, 2020, vol. 19, iss. 1., p. 133-154Fast technological development due to the convergence of calculation power progress, increased storage capacity and advanced network technology makes it possible for companies to collect, process and interlink data in an expanded way. Organizations increasingly tend to use these data for various purposes, such as marketing and personalized services. The European Union's General Data Protection Regulation (hereafter referred to as GDPR) went into effect on 25 May2018. Its privacy regulations apply to any companies collecting or processing personal data in Europe. Many companies had to adjust their data handling processes, privacy policies, and consent forms to comply with the GDPR's transparency requirements. It is important for organizations to ensure that their privacy policies are GDPR compliant. However, it is also important for these policies to be designed with the needs of the human recipient in mind The main problem discussed in this article is the fact that organizations' privacy policies do not comply with GDPR requirements and recommendations of a user-friendly interface. The article introduces the GDPR, including its associated requirements for companies, analyzes privacy policy requirements and best practice for a user- friendly environment. The main purpose of this article is to develop a theoretical privacy policy model under the GDPR, along with a user-friendly interface, the application which could enable organizations to provide relevant content in their privacy policies, ensure consumer rights and encourage greater dissemination of information. This novel theoretical privacy policy model would enable organizational privacy policies to become more user-friendly, more readable and easier to understand, and would bring them into line with the requirements of the GDPR The privacy policy is the main medium of information dissemination between the data controller and the users. Non-compliance with the GDPR poses financial, legal and reputational risks to companies.
15Scopus© Citations 6 General data protection regulation implementation in higher education institutionsItem type:Publication, conference paper[2019][P1a1][S003,S004,S001,N009][8]; EDULEARN19 proceedings: 11th international conference on education and new learning technologies, Palma, Spain. 1-3 July, 2019. Valencia : IATED, 2019. ISBN 9788409120314., p. 2040-2047The General Data Protection Regulation, more commonly known as “GDPR”, is a regulation in the European Union law implemented since the 25th of May, 2018. It aims to increase data protection for EU citizens and individuals within the European Economic Area and simplify international regulations by unifying the European system. The GDPR is a law that protects the personal information of all EU citizens, regardless of where their personal information is located and stored. The GDPR aims to give EU citizens greater protection and control of their personal information. Regulation coverage extends to all organizations – whether or not they have physical EU footprints – that control or process the personal information of EU citizens. Many higher education institutions are gaining international students at a high rate throughout their various online education programs. If any portion of your online educations student base is made up of EU citizens, you will want to ensure that you have GDPR compliance plans in place. The scientific problem, GDPR leaves much to interpretation and higher education institutions unaware of how to implement GDPR requirements. The main purpose of this article is to provide an action plan to help higher education institutions implement GDPR requirements.
14 An analysis of comments containing hate speech: facebook and the delfi news portalItem type:Publication, conference paper[2018][P1a1][S003,S008,N009][7]; EDULEARN 18 : 10th international conference on education and new learning technologies, Palma, Spain, 2nd–4th of July, 2018 : conference proceedings / Edited by L. Gómez Chova, A. López Martínez, I. Candel Torres. Palma : IATED Academy, 2018. ISBN 9788409027095., p. 3413-3419News portals and social networking exert a powerful influence on contemporary society. In order to attract as many readers as possible, news agencies publish the same articles on their websites and on social networking platforms. This is a two-way form of communication where users have the opportunity to express their opinions by commenting on an article in many different news portals and social networks. The comments can add to the users' knowledge and shape their opinions. The problem is that hate speech tends to proliferate in the comments under the articles after they are published. The purpose of the present article is to analyze the peculiarities of hate speech commenting on the Delfi news portal and in the Facebook social network. The academic article presents a limited quantitative and qualitative analysis of Delfi, the most popular Lithuanian news portal, and of the social network known as Facebook, where the author(s) will attempt to identify hateful comments in accordance with certain pre-established criteria and to compare their numbers. The subject matter of the articles published by news portals has a great influence on the number and content of the respective comments. Three topics were chosen for the study: politics, crime and the life of celebrities. The comments chosen for analysis on the Delfi news portal are anonymous, as anonymous commenting is related to freedom of speech, but often promotes hate speech. The article discusses the theoretical aspects of hate speech, examines the options connected to commenting on the Delfi news portal and in the Facebook social network, and evaluates the associated privacy policy. The findings and recommendations of the study have practical implications for reducing hate speech on the Internet.
23 Evaluation of digital shadow economy prevention measuresItem type:Publication, conference paper[2016][T1e][S004,S003,N009][1]; VI International Scientific Conference „21st Century Challenges for Economics and Culture“ : abstracts proceedings. Riga : The University College of Economics and Culture, 2016. ISBN 9789984242019., p. 353 Level and sectors of digital shadow economy: the case of LithuaniaItem type:Publication, research article[2016][S1a][S004,S003,N009][15] ;Gasparėnienė, Ligita; ; Ginevičius, RomualdasEntrepreneurship and sustainability issues. Vilnius : Entrepreneurship and Sustainability Center, 2016, vol. 4, no. 2., p. 183-197The fast pace of g lobal computerization as well as the spread of online activities prompt the changes in the concept of shadow economy. Through cyberspaces, such as social networking platforms, alternative future currency systems, e - commerce, e - business systems or cyber com puter games, real money (or an electronic equivalent of its value) circulates, but in most cases transactions are not account ed and do not generate taxes to the budget of a state. This article is aimed at supplementation of the traditional shadow econom y e stimation methodology with the indicator of digital shadow economy, and performance of the statistical analysis of the data which would allow to identify the areas with the highest scope of digital shadow economy. At present, the scope of shadow economy in Lithuania is estimated by employing one of the two methodologies – developed by the State Tax Inspectorate and by Lithuanian Free Market Institute. It should be noted that neither of these methodologies includes the elements of digital shadow economy. In theoretical level, no methodologies to cover any indicator of digital shadow economy with its relevant features have been developed so far. By employing the method of exp ert evaluation (with participation of 184 experts who work in the Department of Contro l, Lithuanian State Tax Inspectorate), the authors of this article initially estimated the scope of digital shadow economy and identified the areas with the highest scope of digital shadow economy in Lithuania over 2015. The research has revealed that more than a half of the experts have not inspected or identified any subjects illegally operating in e - space. The fact that the scopes of digital shadow economy have not still been estimated indeed burdens the development of any efficient mechanism, which woul d allow to identify and detect illegal economic activities in e - space.
14 Radial basis function collocation method for an elliptic problem with nonlocal multipoint boundary conditionItem type:Publication, research article[2016][S1][N001,N009][9]Engineering analysis with boundary elements. Oxford : Elsevier, 2016, Vol. 67., p. 164-172Radial basis function domain-type collocation method is applied for an elliptic partial differential equation with nonlocal multipoint boundary condition. A geometrically flexible meshless framework is suitable for imposing nonclassical boundary conditions which relate the values of unknown function on the boundary to its values at a discrete set of interior points. Some properties of the method are investigated by a numerical study of a test problem with the manufactured solution. Attention is mainly focused on the influence of nonlocal boundary condition. The standard collocation and least squares approaches are compared. In addition to its geometrical flexibility, the examined method seems to be less restrictive with respect to parameters of nonlocal conditions than, for example, methods based on finite differences.
9Scopus© Citations 7