A5028083798- Topic Modeling
- Natural Language Processing Techniques
- Advanced Text Analysis Techniques
- Sentiment Analysis and Opinion Mining
- Biomedical Text Mining and Ontologies
- Scientific Computing and Data Management
- Fractal and DNA sequence analysis
- Text Readability and Simplification
- Image Retrieval and Classification Techniques
- Culinary Culture and Tourism
- Smart Cities and Technologies
- Nutritional Studies and Diet
- Semantic Web and Ontologies
- Complex Systems and Time Series Analysis
- Machine Learning in Bioinformatics
- Speech and dialogue systems
- Social Media and Politics
- Software Engineering Research
- Meta-analysis and systematic reviews
- Innovative Approaches in Technology and Social Development
- Digital Transformation in Law
- Lexicography and Language Studies
- Explainable Artificial Intelligence (XAI)
- Artificial Intelligence in Healthcare and Education
- Political Conflict and Governance
Athena Research and Innovation Center In Information Communication & Knowledge Technologies
2014-2023
Institute for Language and Speech Processing
2010-2023
Maria Pontiki, Dimitris Galanis, Haris Papageorgiou, Ion Androutsopoulos, Suresh Manandhar, Mohammad AL-Smadi, Mahmoud Al-Ayyoub, Yanyan Zhao, Bing Qin, Orphée De Clercq, Véronique Hoste, Marianna Apidianaki, Xavier Tannier, Natalia Loukachevitch, Evgeniy Kotelnikov, Nuria Bel, Salud María Jiménez-Zafra, Gülşen Eryiğit. Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016). 2016.
SemEval-2015 Task 12, a continuation of SemEval-2014 4, aimed to foster research beyond sentence-or text-level sentiment classification towards Aspect Based Sentiment Analysis.The goal is identify opinions expressed about specific entities (e.g., laptops) and their aspects price).The task provided manually annotated reviews in three domains (restaurants, laptops hotels), common evaluation procedure.It attracted 93 submissions from 16 teams.
We present a method to classify fixed-duration windows of speech as expressing anger or not, which does not require recognition, utterance segmentation, separating the utterances different speakers and can, thus, be easily applied real-world recordings. also introduce task ranking set spoken dialogues by decreasing percentage duration, step towards helping call center supervisors analysts identify conversations requiring further action. Our work is among very few attempts detect emotions in...
We estimate the n-gram entropies of natural language texts in word-length representation and find that these are sensitive to text genre. attribute this sensitivity changes probability distribution lengths single words emphasize crucial role uniformity probabilities having with length between five ten. Furthermore, comparison shuffled data reveals impact word correlations on estimated entropies.
Elisavet Palogiannidi, Athanasia Kolovou, Fenia Christopoulou, Filippos Kokkinos, Elias Iosif, Nikolaos Malandrakis, Haris Papageorgiou, Shrikanth Narayanan, Alexandros Potamianos. Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016). 2016.
Athanasia Kolovou, Filippos Kokkinos, Aris Fergadis, Pinelopi Papalampidi, Elias Iosif, Nikolaos Malandrakis, Elisavet Palogiannidi, Haris Papageorgiou, Shrikanth Narayanan, Alexandros Potamianos. Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017). 2017.
In this paper, we describe a hierarchical bi-directional attention-based Re-current Neural Network (RNN) as reusable sequence encoder architecture, which is used sentence and document for classification. The composed of two RNN equipped with an attention mechanism that identifies captures the most important elements, words or sentences, in followed by dense layer classification task. Our approach utilizes nature documents are sequences sentences words. our model, use word embeddings to...
Classifying scientific publications according to Field-of-Science (FoS) taxonomies is of crucial importance, allowing funders, publishers, scholars, companies and other stakeholders organize literature more effectively. Most existing works address classification either at venue level or solely based on the textual content a research publication. We present SciNoBo, novel system predefined FoS taxonomies, leveraging structural properties publication its citations references organized in...
Science, technology and innovation (STI) policies have evolved in the past decade. We are now progressing towards that more aligned with sustainable development through integrating social, economic environmental dimensions. In this new policy environment, need to keep track of from its conception Science Research has emerged. Argumentation mining, an interdisciplinary NLP field, gives rise required technologies. study, we present first STI-driven multidisciplinary corpus scientific abstracts...
We present1 a personalized ingredient-based Deep Learning recommender on the food domain that exploits ingredients and nutrition information to create recipe representations propose every user more healthier meal. The will be critical component in our Meal Prediction Tool (MPT) designed with focus personalization of services, increasing business efficiency sustainability hospitality, restaurant catering (HoReCa) industry.
Knowledge extraction from scientific literature is a major issue, crucial to promoting transparency, reproducibility, and innovation in the research community. In this work, we present novel approach towards identification, analysis of dataset code/software mentions within literature. We introduce comprehensive dataset, synthetically generated by ChatGPT meticulously curated, augmented, expanded with real snippets text full-text publications Computer Science using human-in-the-loop process....
The volume and velocity of available online sources have changed journalistic research in terms cost effort required for discovering stories. However, the heterogeneity veracity data pose further obstacles knowledge extraction making it a hard task to handle. purpose this study is threefold. Firstly, we present platform automated processing context Computational Journalism. We then propose general methodology event from different sources. Finally, conducted pilot implementation our protest...
The phenomena of climate change transcend all national and regional boundaries. To address this complex challenge, we must determine the areas country interest, in case, Greece, that have been most adversely affected by climate. Greece is surrounded water, a significant part its GDP derived from marine maritime industries, including tourism. Since start IntelComp project, Preparatory Living Lab (PLL) has planned delivered, feeding into development platform on Climate Change Adaptation....
This paper addresses an important problem in example-based machine translation (EBMT), namely how to make retrieval of the example that best matches input more efficient. The use clustering is proposed, enable application same similarity metric first limit search space and then locate available match a database. Evaluation results are presented on large number test cases.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>
SDSN Greece, the Black Sea and Mediterranean, supported by Europe, have established Sustainable Euro-Asian Seas Initiative (SEAs) to accelerate science-driven blue growth SDG implementation in beyond. IntelComp (H2020 project) seeks build an innovative Cloud Platform that will offer AI-based services public administrators policymakers across Europe for data- evidence-driven STI policy design implementation. One of IntelComp's main focus areas is climate change challenge, targeting Blue...