Frosina Koceva | University of Genova (original) (raw)
Papers by Frosina Koceva
English. This paper presents a method for prerequisite learning classification between educationa... more English. This paper presents a method for prerequisite learning classification between educational concepts. The proposed system was developed by adapting a classification algorithm designed for sequencing Learning Objects to the task of ordering concepts from a computer science textbook. In order to apply the system to the new task, for each concept we automatically created a learning unit from the textbook using two criteria based on concept occurrences and burst intervals. Results are promising and suggest that further improvements could highly benefit the results.1 Italiano. Il presente articolo descrive una stategia per l’identificazione di prerequisiti fra concetti didattici. Il sistema proposto è stato realizzato adattando un algoritmo per ordinamento di Learning Objects al compito di ordinamento di concetti estratti da un libro di testo di informatica. Per adeguare il sistema al nuovo scenario, per ogni concetto stata automaticamente creata una unità di apprendimento a parti...
2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS), 2016
Sharing data among applications is a growing phenomenon. With the IoT, this phenomenon becomes mo... more Sharing data among applications is a growing phenomenon. With the IoT, this phenomenon becomes more significant. As already studied in social networks, data sharing has the drawback of privacy risks. Authorization protocols and cryptographic systems may not be enough to ensure that user data and metadata are not used for non-legitimate purposes. There are different scenarios and several personal data management proposals aimed to improve privacy protection. However, a risk that is always present is related to the possibility of processing and aggregating public and authorized data to infer sensitive information and data that the user may not want to share. These approaches, often called inference attacks, concern the disclosure of personal user data and have been widely studied in social networks. In this paper we describe the problem and some techniques to face it, showing its relevance in the IoT. Then we present the concept of an Adaptive Inference Discovery Service AID-S, concei...
Adjunct Publication of the 26th Conference on User Modeling, Adaptation and Personalization, 2018
This paper addresses the design of a model for Question/Answering in an interactive and mobile le... more This paper addresses the design of a model for Question/Answering in an interactive and mobile learning environment. The learner's question can be made through vocal interaction or typed text and the answer is the generation of a personalized learning path. This takes into account the focus and type of the question and some personal features of the learner extracted both from the question and prosodic features, in case of vocal questions. The response is a learning path that preserves the precedence of the prerequisite relations and contains all the relevant concepts for answering the user's question. The main contribution of the paper is to investigate the possibility to exploit educational concept maps in a Q/A interactive learning system.
As well-known, structuring knowledge and digital content has a tremendous potential to enhance me... more As well-known, structuring knowledge and digital content has a tremendous potential to enhance meaningful learning. A straightforward approach is representing key concepts of the subject matter and organizing them in a knowledge structure by means of semantic relations. This results in hypergraphs with typed n-ary relationships, including the so-called prerequisite or propaedeutic relations among concepts. While extracting the whole concept graph from a textbook is our final goal, the focus of this paper is the identification of the propaedeutic relations among concepts. To this aim, we employ a method based on burst analysis and co-occurrence which recognizes, by means of temporal reasoning, prerequisite relations among concepts that share intense periods in the text. The experimental evaluation shows promising results for the extraction of propaedeutic relations without the support of external knowledge.
Adjunct Publication of the 28th ACM Conference on User Modeling, Adaptation and Personalization, 2020
It is our great pleasure to welcome you to the UMAP 2020 LBR and Demo Track, in conjunction with ... more It is our great pleasure to welcome you to the UMAP 2020 LBR and Demo Track, in conjunction with the 28th Conference on User Modelling, Adaptation and Personalization, held online from July 12 to 18, 2020. This track encompasses two categories: (i) Demos, which showcase research prototypes and commercially available products of UMAP-based systems, (ii) Late-breaking Results (LBR), which contain original and unpublished accounts of innovative research ideas, preliminary results, industry showcases, and system prototypes, addressing both the theory and practice of UMAP. The submissions spanned a wide scope of topics, ranging from novel techniques for user and group modeling, to adaptation and personalization implementations across different application scenarios. We received 25 LBR and 4 Demo submissions. Each submission was carefully reviewed by at least 3 members of the Demo and LBR program committee, which consisted of 55 members. Out of this total of 29 submissions, 17 LBR and 2 D...
Adjunct Publication of the 26th Conference on User Modeling, Adaptation and Personalization, 2018
It is our great pleasure to welcome you to the UMAP 2018 IUadaptME workshop. The workshop is focu... more It is our great pleasure to welcome you to the UMAP 2018 IUadaptME workshop. The workshop is focused on two main topics. On the one hand, we are interested in proactive user interfaces that support the user in ubiquitous contexts. This includes applications that can be provided as a service and can be used on any mobile device, with no need for installation or software updates; personalized user interfaces that track the user's progress, adapt to her/his needs and prompt her/him without being disruptive or inconvenient and provide analytics and predictive instruments; smart and personalized information services to explore, search, interact, share, practice and discover qualitative content that fits the user's diverse and changing needs; approaches that are focused on understanding and representing content, users, devices and situations where interaction happens, with special focus on semantics.
Building automatic strategies for organising knowledge contained in textbooks has a tremendous po... more Building automatic strategies for organising knowledge contained in textbooks has a tremendous potential to enhance meaningful learning. Automatic identification of prerequisite relation (PR) between concepts in a textbook is a well-known way for knowledge structuring, yet it is still an open issue. Our research contributes for better understanding and exploring the phenomenon of PR in textbooks, by providing a collection of visualisation techniques for PR exploration and analysis, that we used for the design of and then the refinement of our algorithm for PR extraction.
Digital Libraries present tremendous potential for developing e-learning applications, such as te... more Digital Libraries present tremendous potential for developing e-learning applications, such as text comprehension and question-answering tools. A way to build this kind of tools is structuring the digital content into relevant concepts and dependency relations among them. While the literature offers several approaches for the former, the identification of dependencies, and specifically of prerequisite relations, is still an open issue. We present an approach to manage this task.
This thesis will discuss a knowledge-based model for the design and development of units of learn... more This thesis will discuss a knowledge-based model for the design and development of units of learning and teaching aids. The idea behind this work originates from previous theoretical work on ECM Educational Concept Map (a logical and abstract annotation system, derived from the theories of instructional design), from the open issues in designing instructional authoring system, and from the lack of a well-defined process able to merge pedagogical strategies with systems for the knowledge organization of the domain.
Intelligent textbooks are often engineered with an explicit representation of their concepts and ... more Intelligent textbooks are often engineered with an explicit representation of their concepts and prerequisite relations (PR). PR identification is hence crucial for intelligent textbooks but still presents some challenges, also when performed by human experts. This may cause PRannotated datasets to be inconsistent and compromise the accuracy of automatic creation of enhanced learning materials. This paper investigates possible reasons for PR disagreement and the nature of PR itself, with the aim of contributing to the development of shared strategies for PR annotation, analysis and modelling in textbooks.
This thesis will discuss a knowledge-based model for the design and development of units of learn... more This thesis will discuss a knowledge-based model for the design and development of units of learning and teaching aids. The idea behind this work originates from previous theoretical work on Educational Concept Maps a logical and abstract annotation system derived from the theories of instructional design. Our work is motivated by the open issues in designing instructional authoring system and from the lack of a well-defined process able to merge pedagogical strategies with systems for the knowledge organization of the domain.
2016 11th International Conference for Internet Technology and Secured Transactions (ICITST)
Personal data is an essential component of business models using the Internet of Things (IoT). Ma... more Personal data is an essential component of business models using the Internet of Things (IoT). Massive volumes of personal data are being recorded and analysed about consumers, despite them having limited understanding about how it affects them. Perceptions and preferences in this space influence how consumers choose to interact with the IoT, to a large extent. Yet little is understood about how industry perceives the views of consumers regarding the use of their personal data. To address this gap, we conducted three workshops with IoT industry stakeholders exploring their perspectives of consumer conceptions of the value of personal data in IoT. From the workshops, three overarching analytical themes emerged: (1) A perception of a significant gap between industry and consumers' understanding of what personal data is, who owns it, how it is used in IoT products and how it drives value in IoT businesses; (2) Perceived imbalances of power between industry and consumers in the control of and value extracted from personal data, with implications for inequalities between different consumer groups; and (3) A need for greater education and transparency for consumers, and for industry, about how personal data can be used. We develop a tentative five-point manifesto for the use of personal data in IoT, and conclude that a deeper understanding of consumer perspectives by industry would be positive for the ethical development of the IoT.
Fifth Italian Conference on Computational Linguistics, 2018
Italiano. I coefficienti di Inter-Annotator Agreement sono ampiamente utilizzati in Linguistica C... more Italiano. I coefficienti di Inter-Annotator Agreement sono ampiamente utilizzati in Linguistica Computazionale e NLP per valutare il livello di "affidabilità" delle annotazioni linguistiche. L'articolo propone una breve revisione della letteratura scientifica sull'argomento. English. Agreement indexes are widely used in Computational Linguistics and NLP to assess the reliability of annotation tasks. The paper aims at reviewing the literature on the topic, illustrating chance-corrected coefficients and their interpretation.
The Prerequisite Relation Learning (PRELEARN) task is the EVALITA 2020 shared task on concept pre... more The Prerequisite Relation Learning (PRELEARN) task is the EVALITA 2020 shared task on concept prerequisite learning, which consists of classifying prerequisite relations between pairs of concepts distinguishing between prerequisite pairs and non-prerequisite pairs. Four sub-tasks were defined: two of them define different types of features that participants are allowed to use when training their model, while the other two define the classification scenarios where the proposed models would be tested. In total, 14 runs were submitted by 3 teams comprising 9 total individual participants.
English. This paper presents a method for prerequisite learning classification between educationa... more English. This paper presents a method for prerequisite learning classification between educational concepts. The proposed system was developed by adapting a classification algorithm designed for sequencing Learning Objects to the task of ordering concepts from a computer science textbook. In order to apply the system to the new task, for each concept we automatically created a learning unit from the textbook using two criteria based on concept occurrences and burst intervals. Results are promising and suggest that further improvements could highly benefit the results.1 Italiano. Il presente articolo descrive una stategia per l’identificazione di prerequisiti fra concetti didattici. Il sistema proposto è stato realizzato adattando un algoritmo per ordinamento di Learning Objects al compito di ordinamento di concetti estratti da un libro di testo di informatica. Per adeguare il sistema al nuovo scenario, per ogni concetto stata automaticamente creata una unità di apprendimento a parti...
2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS), 2016
Sharing data among applications is a growing phenomenon. With the IoT, this phenomenon becomes mo... more Sharing data among applications is a growing phenomenon. With the IoT, this phenomenon becomes more significant. As already studied in social networks, data sharing has the drawback of privacy risks. Authorization protocols and cryptographic systems may not be enough to ensure that user data and metadata are not used for non-legitimate purposes. There are different scenarios and several personal data management proposals aimed to improve privacy protection. However, a risk that is always present is related to the possibility of processing and aggregating public and authorized data to infer sensitive information and data that the user may not want to share. These approaches, often called inference attacks, concern the disclosure of personal user data and have been widely studied in social networks. In this paper we describe the problem and some techniques to face it, showing its relevance in the IoT. Then we present the concept of an Adaptive Inference Discovery Service AID-S, concei...
Adjunct Publication of the 26th Conference on User Modeling, Adaptation and Personalization, 2018
This paper addresses the design of a model for Question/Answering in an interactive and mobile le... more This paper addresses the design of a model for Question/Answering in an interactive and mobile learning environment. The learner's question can be made through vocal interaction or typed text and the answer is the generation of a personalized learning path. This takes into account the focus and type of the question and some personal features of the learner extracted both from the question and prosodic features, in case of vocal questions. The response is a learning path that preserves the precedence of the prerequisite relations and contains all the relevant concepts for answering the user's question. The main contribution of the paper is to investigate the possibility to exploit educational concept maps in a Q/A interactive learning system.
As well-known, structuring knowledge and digital content has a tremendous potential to enhance me... more As well-known, structuring knowledge and digital content has a tremendous potential to enhance meaningful learning. A straightforward approach is representing key concepts of the subject matter and organizing them in a knowledge structure by means of semantic relations. This results in hypergraphs with typed n-ary relationships, including the so-called prerequisite or propaedeutic relations among concepts. While extracting the whole concept graph from a textbook is our final goal, the focus of this paper is the identification of the propaedeutic relations among concepts. To this aim, we employ a method based on burst analysis and co-occurrence which recognizes, by means of temporal reasoning, prerequisite relations among concepts that share intense periods in the text. The experimental evaluation shows promising results for the extraction of propaedeutic relations without the support of external knowledge.
Adjunct Publication of the 28th ACM Conference on User Modeling, Adaptation and Personalization, 2020
It is our great pleasure to welcome you to the UMAP 2020 LBR and Demo Track, in conjunction with ... more It is our great pleasure to welcome you to the UMAP 2020 LBR and Demo Track, in conjunction with the 28th Conference on User Modelling, Adaptation and Personalization, held online from July 12 to 18, 2020. This track encompasses two categories: (i) Demos, which showcase research prototypes and commercially available products of UMAP-based systems, (ii) Late-breaking Results (LBR), which contain original and unpublished accounts of innovative research ideas, preliminary results, industry showcases, and system prototypes, addressing both the theory and practice of UMAP. The submissions spanned a wide scope of topics, ranging from novel techniques for user and group modeling, to adaptation and personalization implementations across different application scenarios. We received 25 LBR and 4 Demo submissions. Each submission was carefully reviewed by at least 3 members of the Demo and LBR program committee, which consisted of 55 members. Out of this total of 29 submissions, 17 LBR and 2 D...
Adjunct Publication of the 26th Conference on User Modeling, Adaptation and Personalization, 2018
It is our great pleasure to welcome you to the UMAP 2018 IUadaptME workshop. The workshop is focu... more It is our great pleasure to welcome you to the UMAP 2018 IUadaptME workshop. The workshop is focused on two main topics. On the one hand, we are interested in proactive user interfaces that support the user in ubiquitous contexts. This includes applications that can be provided as a service and can be used on any mobile device, with no need for installation or software updates; personalized user interfaces that track the user's progress, adapt to her/his needs and prompt her/him without being disruptive or inconvenient and provide analytics and predictive instruments; smart and personalized information services to explore, search, interact, share, practice and discover qualitative content that fits the user's diverse and changing needs; approaches that are focused on understanding and representing content, users, devices and situations where interaction happens, with special focus on semantics.
Building automatic strategies for organising knowledge contained in textbooks has a tremendous po... more Building automatic strategies for organising knowledge contained in textbooks has a tremendous potential to enhance meaningful learning. Automatic identification of prerequisite relation (PR) between concepts in a textbook is a well-known way for knowledge structuring, yet it is still an open issue. Our research contributes for better understanding and exploring the phenomenon of PR in textbooks, by providing a collection of visualisation techniques for PR exploration and analysis, that we used for the design of and then the refinement of our algorithm for PR extraction.
Digital Libraries present tremendous potential for developing e-learning applications, such as te... more Digital Libraries present tremendous potential for developing e-learning applications, such as text comprehension and question-answering tools. A way to build this kind of tools is structuring the digital content into relevant concepts and dependency relations among them. While the literature offers several approaches for the former, the identification of dependencies, and specifically of prerequisite relations, is still an open issue. We present an approach to manage this task.
This thesis will discuss a knowledge-based model for the design and development of units of learn... more This thesis will discuss a knowledge-based model for the design and development of units of learning and teaching aids. The idea behind this work originates from previous theoretical work on ECM Educational Concept Map (a logical and abstract annotation system, derived from the theories of instructional design), from the open issues in designing instructional authoring system, and from the lack of a well-defined process able to merge pedagogical strategies with systems for the knowledge organization of the domain.
Intelligent textbooks are often engineered with an explicit representation of their concepts and ... more Intelligent textbooks are often engineered with an explicit representation of their concepts and prerequisite relations (PR). PR identification is hence crucial for intelligent textbooks but still presents some challenges, also when performed by human experts. This may cause PRannotated datasets to be inconsistent and compromise the accuracy of automatic creation of enhanced learning materials. This paper investigates possible reasons for PR disagreement and the nature of PR itself, with the aim of contributing to the development of shared strategies for PR annotation, analysis and modelling in textbooks.
This thesis will discuss a knowledge-based model for the design and development of units of learn... more This thesis will discuss a knowledge-based model for the design and development of units of learning and teaching aids. The idea behind this work originates from previous theoretical work on Educational Concept Maps a logical and abstract annotation system derived from the theories of instructional design. Our work is motivated by the open issues in designing instructional authoring system and from the lack of a well-defined process able to merge pedagogical strategies with systems for the knowledge organization of the domain.
2016 11th International Conference for Internet Technology and Secured Transactions (ICITST)
Personal data is an essential component of business models using the Internet of Things (IoT). Ma... more Personal data is an essential component of business models using the Internet of Things (IoT). Massive volumes of personal data are being recorded and analysed about consumers, despite them having limited understanding about how it affects them. Perceptions and preferences in this space influence how consumers choose to interact with the IoT, to a large extent. Yet little is understood about how industry perceives the views of consumers regarding the use of their personal data. To address this gap, we conducted three workshops with IoT industry stakeholders exploring their perspectives of consumer conceptions of the value of personal data in IoT. From the workshops, three overarching analytical themes emerged: (1) A perception of a significant gap between industry and consumers' understanding of what personal data is, who owns it, how it is used in IoT products and how it drives value in IoT businesses; (2) Perceived imbalances of power between industry and consumers in the control of and value extracted from personal data, with implications for inequalities between different consumer groups; and (3) A need for greater education and transparency for consumers, and for industry, about how personal data can be used. We develop a tentative five-point manifesto for the use of personal data in IoT, and conclude that a deeper understanding of consumer perspectives by industry would be positive for the ethical development of the IoT.
Fifth Italian Conference on Computational Linguistics, 2018
Italiano. I coefficienti di Inter-Annotator Agreement sono ampiamente utilizzati in Linguistica C... more Italiano. I coefficienti di Inter-Annotator Agreement sono ampiamente utilizzati in Linguistica Computazionale e NLP per valutare il livello di "affidabilità" delle annotazioni linguistiche. L'articolo propone una breve revisione della letteratura scientifica sull'argomento. English. Agreement indexes are widely used in Computational Linguistics and NLP to assess the reliability of annotation tasks. The paper aims at reviewing the literature on the topic, illustrating chance-corrected coefficients and their interpretation.
The Prerequisite Relation Learning (PRELEARN) task is the EVALITA 2020 shared task on concept pre... more The Prerequisite Relation Learning (PRELEARN) task is the EVALITA 2020 shared task on concept prerequisite learning, which consists of classifying prerequisite relations between pairs of concepts distinguishing between prerequisite pairs and non-prerequisite pairs. Four sub-tasks were defined: two of them define different types of features that participants are allowed to use when training their model, while the other two define the classification scenarios where the proposed models would be tested. In total, 14 runs were submitted by 3 teams comprising 9 total individual participants.