Tag: iRead4Skills

Regina Duarte Presents Project at World Literacy Summit 2025

Written by iread4skills on April 8, 2025. Posted in News.

This week the city of Oxford is hosting the World Literacy Summit 2025, one of the largest global meetings dedicated to literacy. Today, between 3.40pm and 4pm (local time), the highlight will be the presentation of the iRead4Skills project, led by Regina Duarte, Commissioner of the National Reading Plan 2027 (PNL2027), in room 10 of the Examination School.

The theme of the session, ‘Enhancing literacy for a more inclusive society’, reflects the central aim of the project: to make literacy accessible, functional and aligned with the contemporary demands of the digital society.

The project’s participation in this international forum reinforces its active role in promoting literacy as a tool for inclusion and progress. The World Literacy Summit 2025 brings together delegations from more than 85 countries, with a full programme of debates, workshops and presentations available at: worldliteracysummit.org/programme.

Year 2 Periodic Report of the Project

Written by iread4skills on March 17, 2025. Posted in News.

iRead4Skills has just released its Year 2 Periodic Report, highlighting significant progress in the development of intelligent tools to improve literacy in adults with low reading skills.

Aiming to make reading more accessible and effective for Adult Learning (AL) and Vocational Education and Training (VET) learners, the initiative has been combining artificial intelligence (AI) and natural language processing (NLP) to classify texts according to their complexity and create innovative reading support solutions.

Key Achievements in the Second Year of the Project

📌 Creation of a specialized linguistic corpus in Portuguese, Spanish, and French, classifying texts by complexity levels for further analysis.
📌 Development of the Intelligent Complexity Analyzer (ICA) API, an innovative system that allows evaluating text readability, identifying complex linguistic patterns, and providing recommendations adapted to users’ literacy levels.
📌 Adoption of machine learning models capable of predicting and automatically classifying the reading difficulty of different types of content.
📌 Co-creation workshops with trainers and learners to validate and improve tools, ensuring they meet the real needs of the target audience.
📌 Strengthened collaboration with policymakers and training institutions, ensuring the practical implementation of developed solutions and their integration into educational programs.
📌 Promotion of Open Science, ensuring that collected data and developed methodologies are accessible to researchers and educators.

The Year 2 Periodic Report is now publicly available and can be accessed via the link: https://zenodo.org/records/15038440

iRead4Skills Presented to High School Students in Galicia

Written by iread4skills on March 10, 2025. Posted in News.

On February 20, 2025, the team from the University of Santiago de Compostela (USC), represented by Marcus Garcia Gonzales and Sandra Rodriguez Rey, presented the iRead4Skills project to 12th-grade high school students in Galicia.

During the session, students were introduced to the tools developed as part of the project, aimed at improving reading and writing skills through artificial intelligence. The presentation included a live demonstration of the Text Classifier by Level of Complexity and the Writing Analyzer/Assistant, allowing participants to explore their potential use in education.

This initiative provided an excellent opportunity to engage students in discussions on technology and education, sparking interest in how digital tools can support learning.

The iRead4Skills project remains committed to educational innovation, integrating feedback from teachers and students to refine its applications.

iRead4Skills Co-Creation Workshop

Written by iread4skills on March 10, 2025. Posted in News.

On February 17, 2025, a co-creation workshop was held at NOVA University Lisbon to refine the features and requirements of the applications being developed under the iRead4Skills project: the Text Classifier by Level of Complexity/Difficulty and the Writing Analyzer/Assistant.

The event brought together 12 teachers from various schools and adult education centers who had previously participated in project activities. The session began with a general presentation of iRead4Skills by Raquel Amaro, outlining its objectives and strategy. This was followed by a conceptual overview of the tools presented by Jorge Batista, including a practical demonstration of the Text Classifier, which evaluates text complexity and visually represents vocabulary difficulty.

During the session, guidelines were presented on complexity factors and recommendations for adapting texts to the desired level. Participants also explored the Writing Analyzer’s approach to text evaluation and the different recommendations it provides.

The discussion was highly engaging, with participants raising relevant questions, sharing experiences, and providing constructive feedback. Overall, the tools were well received and considered valuable resources for supporting literacy education. Teachers also emphasized the importance of usability, accessibility, and pedagogical adaptability.

The iRead4Skills co-creation workshop served as a crucial opportunity for dialogue, ensuring that the tools being developed align with the real needs of educators and learners.

Xavier Blanco Presents iRead4Skills at COILES 2025

Written by iread4skills on February 13, 2025. Posted in News.

We are pleased to share that Xavier Blanco (UAB) presented the paper “iRead4Skills. Towards an intelligent system for improving reading skills for the development of fundamental and transversal competences“ at the I Congreso Internacional de Lectura y Escritura (COILES 2025), which took place from 11 to 13 February 2025. This event was organized by the University of Malaga in collaboration with the Alonso Quijano Foundation.

COILES 2025, with its theme “The multidisciplinary development of reading and writing skills“, brought together experts, researchers, and educators to discuss innovative strategies and methodologies in literacy development. The conference provided an excellent platform to showcase iRead4Skills, a pioneering initiative aimed at leveraging intelligent systems to enhance reading competencies and support the acquisition of fundamental and transversal skills.

The iRead4Skills project aligns with the conference’s objective of fostering interdisciplinary approaches to literacy and education, highlighting the role of artificial intelligence and data-driven methodologies in personalized learning experiences. Xavier Blanco’s presentation sparked engaging discussions on the future of digital literacy tools and their potential impact on education.

We are excited to continue contributing to the global conversation on literacy and look forward to further collaborations and knowledge-sharing opportunities.

New iRead4Skills Deliverable: Annotated Corpora by Level of Complexity for FR, PT, and SP

Written by iread4skills on February 5, 2025. Posted in News.

We are pleased to announce the release of Dataset 2: Annotated Corpora by Level of Complexity for French (FR), Portuguese (PT), and Spanish (SP). This dataset is a collection of texts categorized by complexity level and annotated for complexity features, presented in Excel format (.xlsx). The corpora were compiled and annotated under the scope of the iRead4Skills.

Dataset 2 is derived from the previously released Dataset 1: Corpora by Level of Complexity for FR, PT, and SP (DOI: 10.5281/zenodo.10055909), which consists of written texts of various genres and complexity levels. A sample of texts from Dataset 1 was selected for classification and annotation, providing additional data and test sets for the complexity analysis systems in the three project languages.

Data Collection and Annotation Process

The classification and annotation tasks were carried out through a structured methodology:

Texts were distributed to Adult Learning (AL) and Vocational Education Training (VET) Centres, where trainers and students participated in classification tasks.
The classification was conducted via the Qualtrics platform, ensuring a standardized approach.
Participants assigned texts to one of four complexity levels:
- Very Easy (140 texts) – Easily understood by all.
- Easy (140 texts) – Understandable for those with less than 9 years of schooling.
- Plain (140 texts) – Readable at a 9th-grade level.
- More Complex (42 texts) – Challenging for individuals with a 9th-grade education.

For full details on the annotation process, data descriptions, and inter-annotator agreement, refer to the documentation available at Zenodo.

iRead4Skills featured on EPALE – Electronic Platform for Adult Learning in Europe

Written by iread4skills on January 31, 2025. Posted in News.

We are delighted to appear on EPALE, a European, multilingual, open membership community of adult learning professionals, including adult educators and trainers, guidance and support staff, researchers and academics, and policymakers.

EPALE is part of the European Union’s strategy to promote more and better learning opportunities for all adults. The platform supports and strengthens the adult learning professions by enabling members to connect with and learn from colleagues across Europe. It provides a space for knowledge exchange through blog posts, forums, and the Partner Search tool, complemented with physical gatherings. EPALE also offers high-quality, accurate information relevant for adult learning practitioners, fostering engagement and collaboration within the community.

We are pleased to share an insightful interview with Raquel Amaro, the coordinator of the iRead4Skills.

Q&A highlights

What are the goals of this international project?
The iRead4Skills project is dedicated to enhancing adult reading skills by developing an automatic system capable of assessing text complexity. This system will help individuals find suitable reading materials or assist in the creation of such materials. Ultimately, the goal is to bridge skill gaps, improve professional competencies, and enhance access to information and culture for vulnerable and often overlooked populations.

How has it been to work with experts from different nationalities and disciplines?
Collaborating with a diverse, international, and multidisciplinary team has been both rewarding and inspiring. It brings challenges, as different perspectives and expertise must be integrated, but this also results in stronger outcomes. The iRead4Skills project is particularly enriching as it combines knowledge from computational sciences, information technologies, education, social sciences, and humanities.

How can the project contribute to adult literacy improvement across Europe and globally?

Our system is designed to be easily accessible and freely available to trainers, librarians, publishers, and individuals with low literacy skills. By offering tailored reading recommendations, we aim to make reading more engaging and less discouraging. The ability to choose materials that match one’s reading level is crucial for motivation and success, whether in educational settings, informal learning environments, or workplaces. Currently, the system covers Portuguese, French, and Spanish, benefiting speakers of these languages worldwide.

Read the full interview on EPALE: https://epale.ec.europa.eu/en/blog/iread4skills-interview-raquel-amaro

INESC-ID Team of iRead4Skills Earns Best Paper Award at PROPOR 2024

Written by iread4skills on May 13, 2024. Posted in News.

We are delighted to announce that our INESC-ID team members of iRead4Skills, Eugénio Ribeiro, Nuno Mamede, and Jorge Baptista, have been honored with the Best Paper Award at the 16th International Conference on Computational Processing of Portuguese (PROPOR 2024). This prestigious event occurred in Santiago de Compostela, Galiza, on March 12th, 2024, and is renowned as the leading conference focused on Computational Linguistics and Natural Language Processing for the Portuguese language.

The award-winning paper, “Automatic Text Readability Assessment in European Portuguese,” by Eugénio Ribeiro, Nuno Mamede, and Jorge Baptista is accessible in full here.

This recognition underscores the dedication and expertise of our INESC-ID team and their contribution to the field of linguistic technology. We are immensely proud of Eugénio Ribeiro, Nuno Mamede, and Jorge Baptista for their hard work and this well-deserved accolade at one of the most significant gatherings for researchers in Computational Linguistics and Natural Language Processing.

Congratulations to the entire team on this outstanding accomplishment!