Results 1-20 of 98.
((uid:50003041))
![]() Hosseini Kivanani, Nina ![]() ![]() in Hosseini Kivanani, Nina; Vásquez-Correa, Juan Camilo; Schommer, Christoph (Eds.) et al EXPLORING THE USE OF PHONOLOGICAL FEATURES FOR PARKINSON’S DISEASE DETECTION (2023, August) Parkinson’s disease (PD) is a neurodegenerative disorder that causes motor and non-motor symptoms. Speech impairments are one of the early symptoms of PD, but they are not always fully exploited by ... [more ▼] Parkinson’s disease (PD) is a neurodegenerative disorder that causes motor and non-motor symptoms. Speech impairments are one of the early symptoms of PD, but they are not always fully exploited by clinicians. In this study, the use of phonological features extracted from speech data collected from Spanish-speaking patients was explored to predict PD patients from healthy subjects using phonet, which was trained on Spanish data, and PhonVoc, which was trained on English data. These features were then used to train and test several machine learning models. The XGBoost model achieved the best performance in classifying patients from HCs, with an accuracy of over 0.76. However, the model performed better when using a phonological model trained on Spanish data rather than English data. [less ▲] Detailed reference viewed: 21 (0 UL)![]() Lamsiyah, Salima ![]() ![]() in Artificial Intelligence and Machine Learning (2023) Obtaining large-scale and high-quality training data for multi-document summarization (MDS) tasks is time-consuming and resource-intensive, hence, supervised models can only be applied to limited domains ... [more ▼] Obtaining large-scale and high-quality training data for multi-document summarization (MDS) tasks is time-consuming and resource-intensive, hence, supervised models can only be applied to limited domains and languages. In this paper, we introduce unsupervised extractive methods for both generic and query-focused MDS tasks, intending to produce a relevant summary from a collection of documents without using labeled training data or domain knowledge. More specifically, we leverage the potential of transfer learning from recent sentence embedding models to encode the input documents into rich semantic representations. Moreover, we use a coreference resolution system to resolve the broken pronominal coreference expressions in the generated summaries, aiming to improve their cohesion and textual quality. Furthermore, we provide a comparative analysis of several existing sentence embedding models in the context of unsupervised extractive multi-document summarization. Experiments on the standard DUC'2004-2007 datasets demonstrate that the proposed methods are competitive with previous unsupervised methods and are even comparable to recent supervised deep learning-based methods. The empirical results also show that the SimCSE embedding model, based on contrastive learning, achieves substantial improvements over strong sentence embedding models. Finally, the newly involved coreference resolution method is proven to bring a noticeable improvement to the unsupervised extractive MDS task. [less ▲] Detailed reference viewed: 28 (0 UL)![]() Lamsiyah, Salima ![]() ![]() in IEEE Access (2023) Query-Focused Multi-Document Summarization (QF-MDS) is the task of automatically generating a summary from a collection of documents that answers a specific user's query. Extractive methods are primarily ... [more ▼] Query-Focused Multi-Document Summarization (QF-MDS) is the task of automatically generating a summary from a collection of documents that answers a specific user's query. Extractive methods are primarily based on identifying, selecting, and ranking sentences according to their relevance to the given query. These methods have shown promising results; however, they may yield incoherent summaries when pronominal anaphoric expressions appear unbound. To address this issue, this paper proposes a novel method that leverages both contextual embeddings and anaphora resolution methods. More specifically, the Sentence-BERT (SBERT) model is employed to generate contextual embeddings for the sentences in the documents and the user's query. Additionally, the SpanBERT model is utilized to resolve unbound pronominal references in the input sentences of the documents, aiming to improve the cohesiveness of the generated summaries. We have conducted a comprehensive comparative analysis using quantitative and qualitative evaluations against other state-of-the-art systems on the standard DUC'2005 and DUC'2007 datasets. The results obtained show that the proposed method is competitive and outperforms recent query-focused multi-document summarization systems on certain ROUGE evaluation measures. Furthermore, human evaluation results further confirm that our method is able to generate more informative, cohesive, and less redundant summaries. [less ▲] Detailed reference viewed: 29 (1 UL)![]() Lamsiyah, Salima ![]() in Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023) (2023) This paper introduces our participating system to the Explainable Detection of Online Sexism (EDOS) SemEval-2023 - Task 10: Explainable Detection of Online Sexism. The EDOS shared task covers three ... [more ▼] This paper introduces our participating system to the Explainable Detection of Online Sexism (EDOS) SemEval-2023 - Task 10: Explainable Detection of Online Sexism. The EDOS shared task covers three hierarchical sub-tasks for sexism detection, coarse-grained and fine-grained categorization. We have investigated both single-task and multi-task learning based on RoBERTa transformer-based language models. For improving the results, we have performed further pre-training of RoBERTa on the provided unlabeled data. Besides, we have employed a small sample of the unlabeled data for semi-supervised learning using the minimum class-confusion loss. Our system has achieved macro F1 scores of 82.25\textbackslash\%, 67.35\textbackslash\%, and 49.8\textbackslash\% on Tasks A, B, and C, respectively. [less ▲] Detailed reference viewed: 222 (1 UL)![]() Najjar, Amro ![]() ![]() ![]() in XAI: Using Smart Photobooth for Explaining History of Art (2022, December) The rise of Artificial Intelligence has led to advancements in daily life, including applications in industries, telemedicine, farming, and smart cities. It is necessary to have human-AI synergies to ... [more ▼] The rise of Artificial Intelligence has led to advancements in daily life, including applications in industries, telemedicine, farming, and smart cities. It is necessary to have human-AI synergies to guarantee user engagement and provide interactive expert knowledge, despite AI’s success in "less technical" fields. In this article, the possible synergies between humans and AI to explain the development of art history and artistic style transfer are discussed. This study is part of the "Smart Photobooth" project that is able to automatically transform a user’s picture into a well-known artistic style as an interactive approach to introduce the fundamentals of the history of art to the common people and provide them with a concise explanation of the various art painting styles. This study investigates human-AI synergies by combining the explanation produced by an explainable AI mechanism with a human expert’s insights to provide reasons for school students and a larger audience. [less ▲] Detailed reference viewed: 48 (3 UL)![]() Hosseini Kivanani, Nina ![]() ![]() Poster (2022, July) Speakers’ voices are highly individual and for this reason speakers can be identified based on their voice. Nevertheless, voices are often more variable within the same speaker than they are between ... [more ▼] Speakers’ voices are highly individual and for this reason speakers can be identified based on their voice. Nevertheless, voices are often more variable within the same speaker than they are between speakers, which makes it difficult for humans and machines to differentiate between speakers (Hansen, J. H., & Hasan, T., 2015). To date, various machine learning methods have been developed to recognize speakers based on the acoustic characteristics of their speech; however, not all of them have proven equally effective in speaker identification, and depending on the obtained techniques, the system achieves a different result. Here, different machine learning classifiers have been applied to identify the best classification model (i.e., Naïve Bayes (NB), support vector machines (SVM), random forests (RF), & k-nearest neighbors (KNN)) for categorizing 4 speaking styles based on the segment types (voiceless fricatives) considering acoustic features of center of gravity, standard deviation, and skewness. We used a dataset consisting of speech samples from 7 native Persian subjects speaking in 4 different speaking styles: read, spontaneous, clear, and child-directed speech. The results revealed that the best performing model to predict the speakers based on the segment type was RF model with an accuracy of 81,3%, followed by SVM (76.3%), NB (75.4%), and KNN (74%) (Table 1). Our results showed that the RF performed the best for voiceless fricatives /f/, /s/, and / ʃ / which may indicate that these segments are much more speaker-specific than others (Gordon et al., 2002), and the model performance was low for the voiceless fricatives of /h/ and /x/. Performance can be seen in the confusion matrix (Figure 1), which produced high precision and recall values (above 80%) for /f/, /s/ and / ʃ / (Table 2). We found that the model performance improved when the data related to clear speaking style; the information in individual speakers (i.e., voiceless fricatives) are more distinguishable in clear style than other styles (Table 1). [less ▲] Detailed reference viewed: 75 (18 UL)![]() Sirajzade, Joshgun ![]() ![]() ![]() in Applied Informatics, 5th International Conference, ICAI 2022, Arequipa, Peru, October 27–29, 2022, Proceedings (2022) In this paper we investigate how scientific and medical papers about Covid-19 can be effectively mined. For this purpose we use the CORD19 dataset which is a huge collection of all papers published about ... [more ▼] In this paper we investigate how scientific and medical papers about Covid-19 can be effectively mined. For this purpose we use the CORD19 dataset which is a huge collection of all papers published about and around the SARS-CoV2 virus and the pandemic it caused. We discuss how classical text mining algorithms like Latent Semantic Analysis (LSA) or its modern version Latent Drichlet Allocation (LDA) can be used for this purpose and also touch more modern variant of these algorithms like word2vec which came with deep learning wave and show their advantages and disadvantages each. We finish the paper with showing some topic examples from the corpus and answer questions such as which topics are the most prominent for the corpus or how many percentage of the corpus is dedicated to them. We also give a discussion of how topics around RNA research in connection with Covid-19 can be examined. [less ▲] Detailed reference viewed: 47 (3 UL)![]() Leiva, Luis A. ![]() ![]() ![]() Book published by Springer (2022) Detailed reference viewed: 82 (1 UL)![]() Schommer, Christoph ![]() Speeches/Talks (2021) From a technical point of view, one can say that there have been extraordinary developments in the broad areas of Artificial Intelligence and Data Science that increasingly determine our lives. Think, for ... [more ▼] From a technical point of view, one can say that there have been extraordinary developments in the broad areas of Artificial Intelligence and Data Science that increasingly determine our lives. Think, for example, of the many intelligent assistants that are playing an ever greater role in road traffic, in hospitals and in many other areas of application. Art is affected by this as well, because the computer-assisted and AI-supported implementation of creative ideas and thoughts and experimentation with new things have led to innovative results. [less ▲] Detailed reference viewed: 128 (6 UL)![]() Schommer, Christoph ![]() Speeches/Talks (2021) The present and the future will be determined by topics such as artificial intelligence and the question arises to what extent an often quoted usefulness, an "AI for Social Good" and "AI is for Humans ... [more ▼] The present and the future will be determined by topics such as artificial intelligence and the question arises to what extent an often quoted usefulness, an "AI for Social Good" and "AI is for Humans" really applies. The lecture is intended to provide food for thought. [less ▲] Detailed reference viewed: 83 (4 UL)![]() Schommer, Christoph ![]() Speeches/Talks (2021) This talk, which was part of a 2-hour panel and delivered as a 20-minute presentation without slides, came at the invitation of the Luxembourg Embassy in Brussels, Belgium. The speech is a statement and ... [more ▼] This talk, which was part of a 2-hour panel and delivered as a 20-minute presentation without slides, came at the invitation of the Luxembourg Embassy in Brussels, Belgium. The speech is a statement and part of the International Symposium "The Future of Living", supported by EUNIC - EU National Institutes of Culture. The event was under the patronage of the Slovenian Presidency of the European Parliament. In the lecture, I briefly presented some selected ideas, observations, and examples on the symposium topic as well as my view of a future living together. [less ▲] Detailed reference viewed: 159 (5 UL)![]() Schommer, Christoph ![]() Speeches/Talks (2021) Keynote Talk "Getting Creative - AI and Arts"; AIFA - Artificial Intelligence and the Future of Arts Detailed reference viewed: 153 (7 UL)![]() Schommer, Christoph ![]() Speeches/Talks (2021) La Commission Luxembourgeoise pour la coopération avec l’UNESCO et la Bibliothèque nationale du Luxembourg invitent à leur cycle de conférences dans le cadre des « Rendez-Vous de l’UNESCO ». Le thème ... [more ▼] La Commission Luxembourgeoise pour la coopération avec l’UNESCO et la Bibliothèque nationale du Luxembourg invitent à leur cycle de conférences dans le cadre des « Rendez-Vous de l’UNESCO ». Le thème central du cycle est cette année : « Mankind and Media. Rethinking the Roles in the Age of Information » Les dernières années ont en outre montré clairement qu’on est fortement dépendant de la transmission des informations vite et performante : on demande toujours plus de données, plus d’informations, plus de connectivité. Mais peut-on conserver une vue d’ensemble dans ce flux d’information hyperrapide ? Peut-on encore contrôler la transformation des informations par les médias numériques ? Où en est l’humain devant l’évolution exponentielle des technologies, de l’intelligence artificielle et du commerce de nos données ? Qui reste à l’origine de l’information et qui comment nous l’interprétons ? Les conférences traitent cette thématique en différentes perspectives. Le 25 février, Prof. Dr. Christoph Schommer de l’Université du Luxembourg fera l’ouverture en posant la question : L’intelligence Artificielle? Où en sommes-nous ? Il abordera les possibilités, chances et risques de ces nouvelles technologies. 4 journalistes se réuniront le 22 avril et discuteront, animés par Josée Hansen, la transmission d’informations et l’objectivité dans le temps des « Fake News » et du « Click Bait ». Dr. Manuela di Franco nous introduira le 1er juillet dans le monde des dessins animés et se penchera sur leur rôle dans la transmission des messages subliminaux et propagandistes. Ian di Toffoli et Prof. Dr. Lukas K. Sosoe s’entretiendront le 30 septembre sur les problèmes et limites de l’éthique dans le monde digital. Qu’il ne faut pas oublier l’art dans la transmission des messages nous rappelleront les 4 artistes qui poursuivront le 2nd décembre, dans la dernière table ronde de ce cycle modérée par Dr. Nora Schleich, le rôle de la liberté artistique face à la libre expression des opinions. Que peut et même doit faire l’art ? La Bibliothèque nationale du Luxembourg aménagera pour chacune de ces soirées, qui auront lieu toujours jeudi à 19hrs, une salle assez grande pour assurer la réunion en présence physique et, le cas échéant, en respect des restrictions actuelles. 25.02.2021 Der Mensch und die Künstliche Intelligenz. Wo stehen wir? (DE) Un entretien avec Prof. Christoph Schommer, modéré par Dr. Nora Schleich, traduit en anglais 22.04.2021 Mediepluralismus – Eng Garantie fir Demokratie am 21. Joerhonnert? (LU) Une table-ronde avec François Aulner, Pia Oppel, Christoph Bumb et Jean-Louis Siweck, modérée par Josée Hansen 01.07.2021 Transmission of Information via Specific Media – Propagandistic Messages in Comics (EN) Une présentation de Dr. Manuela di Franco, traduit en français 30.09.2021 L’Ethique dans la société de l’Information (FR) Un discours avec Prof. Lukas Sosoe, modéré par Ian de Toffoli, traduit en anglais 02.12.2021 Artistesch Fräiheet a Meenungsfräiheet (LU) Une table-ronde avec Justine Blau, Dr. Cédric Kayer, Filip Markiewicz et Anne Simon., modérée par Dr. Nora Schleich Comme les places sont limitées, il faudra s’inscrire au plus tard trois jours avant la conférence en envoyant un courriel à : reservation@bnl.etat.lu Pour plus d’informations, veuillez accéder le site internet : www.unesco.lu; www.bnl.lu [less ▲] Detailed reference viewed: 211 (4 UL)![]() Leiva, Luis A. ![]() ![]() ![]() Book published by BnL (2021) Detailed reference viewed: 243 (33 UL)![]() Schommer, Christoph ![]() ![]() ![]() Scientific Conference (2021) The research field between Artificial Intelligence and Health sciences has established itself as a central research direction in recent years and has also further increased social interest. On the one ... [more ▼] The research field between Artificial Intelligence and Health sciences has established itself as a central research direction in recent years and has also further increased social interest. On the one hand, this is due to the emergence of medical mass data and their use for AI-related fields, such as machine learning, human-computer interfaces and natural language-processing systems, and on the other hand, it is also due to the steadily growing social interest, which is not determined by the current Covid 19 pandemic. To this end, the lecture series is intended to provide an opportunity for scientific exchange. [less ▲] Detailed reference viewed: 80 (3 UL)![]() Schommer, Christoph ![]() Speeches/Talks (2020) Detailed reference viewed: 158 (3 UL)![]() ; Schommer, Christoph ![]() in Sharma, Rudrani (Ed.) Building up Explainability in Multi-layer Perceptrons for Credit Risk Modeling (2020, October 09) Granting loans is one of the major concerns of financial institutions due to the risks of default borrowers. Default prediction by the neural networks is a popular technique for credit risk modeling ... [more ▼] Granting loans is one of the major concerns of financial institutions due to the risks of default borrowers. Default prediction by the neural networks is a popular technique for credit risk modeling. Neural networks generally offer the accurate predictions that help banks to prevent financial losses and grow their business by approving more creditworthy borrowers. Although neural networks are capable of capturing the complex, non-linear relationships between a large number of features and output, these models act as black boxes. This is a graduation project paper that is focused on loan default risk prediction by multi-layer perceptron neural network and building up explainability to some degree in the trained neural networks through sensitivity analysis. The architecture of a multi-layer perceptron neural network with the best result is used to help the credit-risk manager in explaining why an applicant is a defaulter or non-defaulter. The prediction of a trained multi-layer perceptron neural network is explained by mapping input features and target variables directly using a model-agnostic explanation as well as a modelspecific explanation. Lastly, a comparison is performed between two explanation methods. [less ▲] Detailed reference viewed: 111 (5 UL)![]() Kamlovskaya, Ekaterina ![]() ![]() Scientific Conference (2020, September 29) This submission presents intermediate results of the PhD project analysing discourses in a corpus of Australian Aboriginal autobiographies with word embedding modelling. Detailed reference viewed: 68 (12 UL)![]() Sirajzade, Joshgun ![]() ![]() ![]() in Besacier, Laurent; Sakti, Sakriani; Soria, Claudia (Eds.) et al Proceedings of the LREC 2020 1st Joint SLTU and CCURL Workshop (SLTU-CCURL 2020) (2020, May) The aim of this paper is to present a framework developed for crowdsourcing sentiment annotation for the low-resource language Luxembourgish. Our tool is easily accessible through a web interface and ... [more ▼] The aim of this paper is to present a framework developed for crowdsourcing sentiment annotation for the low-resource language Luxembourgish. Our tool is easily accessible through a web interface and facilitates sentence-level annotation of several annotators in parallel. In the heart of our framework is an XML database, which serves as central part linking several components. The corpus in the database consists of news articles and user comments. One of the components is LuNa, a tool for linguistic preprocessing of the data set. It tokenizes the text, splits it into sentences and assigns POS-tags to the tokens. After that, the preprocessed text is stored in XML format into the database. The Sentiment Annotation Tool, which is a browser-based tool, then enables the annotation of split sentences from the database. The Sentiment Engine, a separate module, is trained with this material in order to annotate the whole data set and analyze the sentiment of the comments over time and in relationship to the news articles. The gained knowledge can again be used to improve the sentiment classification on the one hand and on the other hand to understand the sentiment phenomenon from the linguistic point of view. [less ▲] Detailed reference viewed: 157 (29 UL)![]() Sirajzade, Joshgun ![]() ![]() ![]() in Beermann, Dorothee; Besacier, Laurent; Sakti, Sakriani (Eds.) et al Proceedings of the LREC 2020 1st Joint SLTU and CCURL Workshop(SLTU-CCURL 2020) (2020, May) The aim of this paper is to investigate the role of Luxembourgish adjectives in expressing sentiments in user comments written at the web presence of rtl.lu (RTL is the abbreviation for Radio Television ... [more ▼] The aim of this paper is to investigate the role of Luxembourgish adjectives in expressing sentiments in user comments written at the web presence of rtl.lu (RTL is the abbreviation for Radio Television Lëtzebuerg). Alongside many textual features or representations, adjectives could be used in order to detect sentiment, even on a sentence or comment level. In fact, they are also by themselves one of the best ways to describe a sentiment, despite the fact that other word classes such as nouns, verbs, adverbs or conjunctions can also be utilized for this purpose. The empirical part of this study focuses on a list of adjectives that were extracted from an annotated corpus. The corpus contains the part of speech tags of individual words and sentiment annotation on the adjective, sentence, and comment level. Suffixes of Luxembourgish adjectives like -esch, -eg, -lech, -al, -el, -iv, -ent, -los, -bar and the prefix on- were explicitly investigated, especially by paying attention to their role in regards to building a model by applying classical machine learning techniques. We also considered the interaction of adjectives with other grammatical means, especially other part of speeches, e.g. negations, which can completely reverse the meaning, thus the sentiment of an utterance. [less ▲] Detailed reference viewed: 151 (21 UL) |
||