The paper presents the main results of a survey aimed at a comparative analysis of some widely us... more The paper presents the main results of a survey aimed at a comparative analysis of some widely used open source software tools for creating digital repositories with respect to their applicability for web publishing of digitized Bulgarian cultural heritage collections and making them accessible via Europeana. The parallel has been made according to a well-defined set of features like availability of convenient authoring/annotation and browsing tools, image scaling and high resolution support, functionalities of the built-in search engine, recognizable file types, user management, OAI-PMH compliance, etc. The Europeana Data Model and the Europeana Semantic Elements standard are briefly described from the viewpoint of publishing in Europeana. Some practical recommendations and conclusions are given as a final result.
Abstract. The paper presents results from the development of a methodology and corresponding soft... more Abstract. The paper presents results from the development of a methodology and corresponding software tools for building an academic repository. The repository was filled up with gaming material. The repository architecture and key features of the search engine are discussed. The emphasis falls on solutions of the large set of problems concerning the development of proper mechanisms for semantics-based search in a digital repository. 1. Introduction. The
The paper presents in brief a methodology for development of tools for knowledge-based search in ... more The paper presents in brief a methodology for development of tools for knowledge-based search in repositories of digitized manuscripts. It is designated to assist the search activities in collections that may enlist XML documents which should be catalogue descriptions or marked-up full texts of mediaeval manuscripts. The suggested methodology is directed to the development of software environments that will be able to deal with relatively complex user queries containing words or phrases that are considered as domain concepts. The emphasis in this methodology falls on two main types of activities: development of proper ontologies describing the conceptual knowledge relevant to the chosen domain(s) and development of proper intelligent agents for search and processing purposes that are able to retrieve and filter documents by their semantic properties. Some considerations related to the implementation of our methodology are presented as well. The first version of a software tool for k...
Semantic data enrichment adds information to raw data to allow computational reasoning based on t... more Semantic data enrichment adds information to raw data to allow computational reasoning based on the meaning of data. With the introduction of Linked Data a lot of work is spent on combining existing tools for specific enhancement needs into multi-domain reusable enhancement pipeline.
Artificial Intelligence: Methodology, Systems, and Applications
The use of machine learning in disease prediction and prognosis is part of a growing trend of per... more The use of machine learning in disease prediction and prognosis is part of a growing trend of personalized and predictive medicine. Cancer studies are domain of active machine learning implementation in particular in sense of accuracy of cancer prognosis and prediction. The accuracy of survival time prediction in breast cancer is the main object of the study. Two major features for survival time prediction, based on clinical data are used: the created in the study tumor integrated clinical feature and Nottingham prognostic index. The applied machine learning methods aside with data normalisation and classification provide promising results for accuracy of survival time prediction. Results showed prepotency of the support vector regression modles - linear and decision tree regression models, for more accurate prediction of the survival time in breast cancer. Cross-validation, based on four parameters for error evaluation, confirms the results of the model performance concerning the accuracy of survival time prediction in breast cancer.
The paper discusses some aspects of a project aimed at the development of a methodology and prope... more The paper discusses some aspects of a project aimed at the development of a methodology and proper software tools for building academic digital libraries. A particular functional model of academic digital library has been proposed and analyzed. The emphasis of the presentation falls on some solutions of the large set of problems concerning the development of adequate mechanisms for semantics oriented search in multilingual digital libraries. An ontology-based approach is suggested in order to standardize the semantic annotation of the library resources and to facilitate the implementation of the functionality of the search engine. A proper subject ontology covering the area of Computer Science has been developed for this purpose. The paper discusses the requirements of the basic types of users of an academic digital library and suggests some relevant solutions.
The paper presents a work in progress directed to the creation of a semantic digital repository o... more The paper presents a work in progress directed to the creation of a semantic digital repository of scholarly resources in the area of bioinformatics. A special attention has been paid to the design of a lightweight subject ontology that will be used for automated conceptual annotation of the dynamically entering new materials and as a knowledge source for intelligent search over the repository. Some issues concerning the implementation of the repository and a suitable tool for search and document retrieval over its resources are discussed as well.
In today’s world of digital transformation and big data the issues related to the semantic intero... more In today’s world of digital transformation and big data the issues related to the semantic interoperability of different information systems that support the decision making activities in the same or similar areas are becoming increasingly important. It is therefore natural for such issues to be in the focus of the education at master’s level in a variety of professional fi elds. The paper discusses the experience in this regard of the Master’s program in Artifi cial Intelligence at the Faculty of Mathematics and Informatics at Sofi a University, focusing on some good examples of student projects.
The paper discusses some aspects of an ongoing project aimed at the development of a methodology ... more The paper discusses some aspects of an ongoing project aimed at the development of a methodology and proper software tools for building and usage of academic digital libraries. A particular functional model of academic digital library has been proposed and analyzed. The emphasis falls on some solutions of the large set of problems concerning the development of adequate mechanisms for semantics-oriented search in multilingual digital libraries. An ontology-based approach is suggested in order to standardize the semantic annotation of the library resources and to facilitate the implementation of the functionality of the search engine. The main features of a prototype of knowledge-based search engine for a multilingual academic digital library with research and learning materials are discussed. This search engine uses proper ontologies describing the conceptual knowledge considerable for the chosen domains and in this way it is capable of retrieving and filtering documents by their sem...
The paper presents the main aspects of an ongoing project aimed at the development of technologie... more The paper presents the main aspects of an ongoing project aimed at the development of technologies for digitization of Bulgarian folk music and building a heterogeneous digital library with Bulgarian folk songs presented with their music, notes and text. The digitization of sound recordings and notations of folk music will facilitate the access to them, will provide their preservation in another quality and will present the opportunity for these recordings and notations to be used by a greater number of users through webbased access. The digital library will provide both digital preservation of the sound recordings and notations of Bulgarian folk songs and a possibility for new interpretations of the archaic Bulgarian folklore heritage. Information technologies for presentation of Bulgarian folk songs is an interdisciplinary project, which imposes a specific research approach, combining the plurality and dialogue of a multitude of viewpoints, positions and perspectives.
The paper discusses several aspects of the use of ontological knowledge and some concomitant Se- ... more The paper discusses several aspects of the use of ontological knowledge and some concomitant Se- mantic Web technologies in the development of software tools for operative access to digitized cultural heritage and archival collections. The emphasis falls on a number of general issues like semantic mark-up of content, information integration, interoperability of ontologies etc. Some domain-specific problems e.g. the scope of the ontologies that are needed for the purpose (and which ones should the heritage sector develop and which ones will be possible to borrow from other sectors) have been also analyzed. Two successful projects directed to the implementation of ontology-driven access to various types of cultural heritage repositories have been analyzed as examples of best practices in the area. The experience in building software tools for knowledge-based search in repositories of digitized manuscripts and archival materials gained at the Institute of Mathematics and Infor- matics ...
Ontology alignment is regarded as one of the core tasks in many Web services. It is concerned wit... more Ontology alignment is regarded as one of the core tasks in many Web services. It is concerned with finding the correspondences between separate ontologies by identifying concepts with the same or similar semantics in order to resolve semantic heterogeneity between them. Existing ontology alignment techniques are tailored towards today’s ontology languages, which are not capable of representing and reasoning with uncertain or incomplete information. It is expected, however, that future Semantic Web services will rely on the development and use of proper domain ontologies. Alignment of such ontologies goes beyond standard concept matching, and requires nonstandard logic processing. In this paper, we present an alignment technique utilizing an alternative, rule-based representation, which provides a uniform framework for representing and mapping heterogeneous ontologies. To justify and illustrate our research, we describe an example application scenario.
2020 International Conference Automatics and Informatics (ICAI), 2020
Along with the massive influence of computational technologies in medical research and practice, ... more Along with the massive influence of computational technologies in medical research and practice, the wide generation of patient, clinical and lab tests data make the assistance of intelligent information systems very important for correct therapy, surveillance and advising of the patient. Medical data is highly heterogeneous and with different structure and formats. Principal tasks of the work are focused at intelligent integration of all data related to a particular patient. Aside with the methodological aspects of linking medical data, an important part of the discussed study focuses on the design and development of a special purpose decision support system. This system enriches a specific ontology for diabetes mellitus with a set of newly defined rules, supporting the development of broad and more precise personalized dietary recommendations as a part of digitized healthcare assistance.
Data analysis after parallel sequencing is a process that uses combinations of software tools tha... more Data analysis after parallel sequencing is a process that uses combinations of software tools that is often subject to experimentation and on-the-fly substitution, with the necessary file conversion. This article presents a developing system for creating and managing workflows aiding the tasks one encounters after parallel sequences, particularly in the area of metagenomics. The semantics, description language and software implementation aim to allow the creation of flexible, configurable workflows that are suitable for sharing and are easy to manipulate through software or by hand. The execution system design provides user-defined operations and interchangeability between an operation and a workflow. This allows significant extensibility, which can be further complemented with distributed computing and remote management interfaces.
The aim of this paper is to review some methods for text authorship attribution and to discuss th... more The aim of this paper is to review some methods for text authorship attribution and to discuss the development of a software library with tools for automatic authorship attribution. The presentation is focused on an analysis of two groups of tools oriented to: (1) methods for extraction of features and (2) methods for computing the distance between character strings based on data compression algorithms.
The paper analyzes the most signifi cant issues related to the information extraction from docume... more The paper analyzes the most signifi cant issues related to the information extraction from documents and discusses some theoretical models suitable for recognition. It presents, in brief, the main components of a hybrid approach, which combines an application of a domain ontology and deep learning techniques to recognize and classify the document structure effi ciently. The approach is based on the use of a symbolic logic inference model to utilize the knowledge about the document semantics. The document’s segments are referring to the conceptual purpose of the document, which is recognized by applying modern architecture for object detection. An appropriate domain ontology solves the issues related to the semantic completeness of the document’s data. Several experiments have been carried out and the results obtained have been analyzed in terms of the applicability of various modern technologies for the implementation of document understanding system.
The present work presents a survey of the present state of programming languages employing type s... more The present work presents a survey of the present state of programming languages employing type systems with dependent types, i.e., types parameterized by values. The features of the languages are reviewed and compared with an emphasis on the usefulness of the type checker as a tool for asserting correctness statements about programs. Keywords: type systems, dependent types, programming languages, program correctness
The paper presents results from the development of a methodology and corresponding software tools... more The paper presents results from the development of a methodology and corresponding software tools for building an academic repository. The repository was filled up with gaming material. The repository architecture and key features of the search engine are discussed. The emphasis falls on solutions of the large set of problems concerning the development of proper mechanisms for semantics-based search in a digital repository.
The paper presents a methodology for development of tools for semantics oriented search in reposi... more The paper presents a methodology for development of tools for semantics oriented search in repositories of digitized manuscripts. This methodology is based on the application of some Semantic Web technologies to existing manuscript collections that may include: electronic catalogue containing marked-up manuscript descriptions, full texts of manuscripts, digital images of manuscript pages. It is directed to the development of software environments that will be able to deal with complex user queries and answer questions using ontological knowledge and dictionaries of synonyms.
The paper presents the main results of a survey aimed at a comparative analysis of some widely us... more The paper presents the main results of a survey aimed at a comparative analysis of some widely used open source software tools for creating digital repositories with respect to their applicability for web publishing of digitized Bulgarian cultural heritage collections and making them accessible via Europeana. The parallel has been made according to a well-defined set of features like availability of convenient authoring/annotation and browsing tools, image scaling and high resolution support, functionalities of the built-in search engine, recognizable file types, user management, OAI-PMH compliance, etc. The Europeana Data Model and the Europeana Semantic Elements standard are briefly described from the viewpoint of publishing in Europeana. Some practical recommendations and conclusions are given as a final result.
Abstract. The paper presents results from the development of a methodology and corresponding soft... more Abstract. The paper presents results from the development of a methodology and corresponding software tools for building an academic repository. The repository was filled up with gaming material. The repository architecture and key features of the search engine are discussed. The emphasis falls on solutions of the large set of problems concerning the development of proper mechanisms for semantics-based search in a digital repository. 1. Introduction. The
The paper presents in brief a methodology for development of tools for knowledge-based search in ... more The paper presents in brief a methodology for development of tools for knowledge-based search in repositories of digitized manuscripts. It is designated to assist the search activities in collections that may enlist XML documents which should be catalogue descriptions or marked-up full texts of mediaeval manuscripts. The suggested methodology is directed to the development of software environments that will be able to deal with relatively complex user queries containing words or phrases that are considered as domain concepts. The emphasis in this methodology falls on two main types of activities: development of proper ontologies describing the conceptual knowledge relevant to the chosen domain(s) and development of proper intelligent agents for search and processing purposes that are able to retrieve and filter documents by their semantic properties. Some considerations related to the implementation of our methodology are presented as well. The first version of a software tool for k...
Semantic data enrichment adds information to raw data to allow computational reasoning based on t... more Semantic data enrichment adds information to raw data to allow computational reasoning based on the meaning of data. With the introduction of Linked Data a lot of work is spent on combining existing tools for specific enhancement needs into multi-domain reusable enhancement pipeline.
Artificial Intelligence: Methodology, Systems, and Applications
The use of machine learning in disease prediction and prognosis is part of a growing trend of per... more The use of machine learning in disease prediction and prognosis is part of a growing trend of personalized and predictive medicine. Cancer studies are domain of active machine learning implementation in particular in sense of accuracy of cancer prognosis and prediction. The accuracy of survival time prediction in breast cancer is the main object of the study. Two major features for survival time prediction, based on clinical data are used: the created in the study tumor integrated clinical feature and Nottingham prognostic index. The applied machine learning methods aside with data normalisation and classification provide promising results for accuracy of survival time prediction. Results showed prepotency of the support vector regression modles - linear and decision tree regression models, for more accurate prediction of the survival time in breast cancer. Cross-validation, based on four parameters for error evaluation, confirms the results of the model performance concerning the accuracy of survival time prediction in breast cancer.
The paper discusses some aspects of a project aimed at the development of a methodology and prope... more The paper discusses some aspects of a project aimed at the development of a methodology and proper software tools for building academic digital libraries. A particular functional model of academic digital library has been proposed and analyzed. The emphasis of the presentation falls on some solutions of the large set of problems concerning the development of adequate mechanisms for semantics oriented search in multilingual digital libraries. An ontology-based approach is suggested in order to standardize the semantic annotation of the library resources and to facilitate the implementation of the functionality of the search engine. A proper subject ontology covering the area of Computer Science has been developed for this purpose. The paper discusses the requirements of the basic types of users of an academic digital library and suggests some relevant solutions.
The paper presents a work in progress directed to the creation of a semantic digital repository o... more The paper presents a work in progress directed to the creation of a semantic digital repository of scholarly resources in the area of bioinformatics. A special attention has been paid to the design of a lightweight subject ontology that will be used for automated conceptual annotation of the dynamically entering new materials and as a knowledge source for intelligent search over the repository. Some issues concerning the implementation of the repository and a suitable tool for search and document retrieval over its resources are discussed as well.
In today’s world of digital transformation and big data the issues related to the semantic intero... more In today’s world of digital transformation and big data the issues related to the semantic interoperability of different information systems that support the decision making activities in the same or similar areas are becoming increasingly important. It is therefore natural for such issues to be in the focus of the education at master’s level in a variety of professional fi elds. The paper discusses the experience in this regard of the Master’s program in Artifi cial Intelligence at the Faculty of Mathematics and Informatics at Sofi a University, focusing on some good examples of student projects.
The paper discusses some aspects of an ongoing project aimed at the development of a methodology ... more The paper discusses some aspects of an ongoing project aimed at the development of a methodology and proper software tools for building and usage of academic digital libraries. A particular functional model of academic digital library has been proposed and analyzed. The emphasis falls on some solutions of the large set of problems concerning the development of adequate mechanisms for semantics-oriented search in multilingual digital libraries. An ontology-based approach is suggested in order to standardize the semantic annotation of the library resources and to facilitate the implementation of the functionality of the search engine. The main features of a prototype of knowledge-based search engine for a multilingual academic digital library with research and learning materials are discussed. This search engine uses proper ontologies describing the conceptual knowledge considerable for the chosen domains and in this way it is capable of retrieving and filtering documents by their sem...
The paper presents the main aspects of an ongoing project aimed at the development of technologie... more The paper presents the main aspects of an ongoing project aimed at the development of technologies for digitization of Bulgarian folk music and building a heterogeneous digital library with Bulgarian folk songs presented with their music, notes and text. The digitization of sound recordings and notations of folk music will facilitate the access to them, will provide their preservation in another quality and will present the opportunity for these recordings and notations to be used by a greater number of users through webbased access. The digital library will provide both digital preservation of the sound recordings and notations of Bulgarian folk songs and a possibility for new interpretations of the archaic Bulgarian folklore heritage. Information technologies for presentation of Bulgarian folk songs is an interdisciplinary project, which imposes a specific research approach, combining the plurality and dialogue of a multitude of viewpoints, positions and perspectives.
The paper discusses several aspects of the use of ontological knowledge and some concomitant Se- ... more The paper discusses several aspects of the use of ontological knowledge and some concomitant Se- mantic Web technologies in the development of software tools for operative access to digitized cultural heritage and archival collections. The emphasis falls on a number of general issues like semantic mark-up of content, information integration, interoperability of ontologies etc. Some domain-specific problems e.g. the scope of the ontologies that are needed for the purpose (and which ones should the heritage sector develop and which ones will be possible to borrow from other sectors) have been also analyzed. Two successful projects directed to the implementation of ontology-driven access to various types of cultural heritage repositories have been analyzed as examples of best practices in the area. The experience in building software tools for knowledge-based search in repositories of digitized manuscripts and archival materials gained at the Institute of Mathematics and Infor- matics ...
Ontology alignment is regarded as one of the core tasks in many Web services. It is concerned wit... more Ontology alignment is regarded as one of the core tasks in many Web services. It is concerned with finding the correspondences between separate ontologies by identifying concepts with the same or similar semantics in order to resolve semantic heterogeneity between them. Existing ontology alignment techniques are tailored towards today’s ontology languages, which are not capable of representing and reasoning with uncertain or incomplete information. It is expected, however, that future Semantic Web services will rely on the development and use of proper domain ontologies. Alignment of such ontologies goes beyond standard concept matching, and requires nonstandard logic processing. In this paper, we present an alignment technique utilizing an alternative, rule-based representation, which provides a uniform framework for representing and mapping heterogeneous ontologies. To justify and illustrate our research, we describe an example application scenario.
2020 International Conference Automatics and Informatics (ICAI), 2020
Along with the massive influence of computational technologies in medical research and practice, ... more Along with the massive influence of computational technologies in medical research and practice, the wide generation of patient, clinical and lab tests data make the assistance of intelligent information systems very important for correct therapy, surveillance and advising of the patient. Medical data is highly heterogeneous and with different structure and formats. Principal tasks of the work are focused at intelligent integration of all data related to a particular patient. Aside with the methodological aspects of linking medical data, an important part of the discussed study focuses on the design and development of a special purpose decision support system. This system enriches a specific ontology for diabetes mellitus with a set of newly defined rules, supporting the development of broad and more precise personalized dietary recommendations as a part of digitized healthcare assistance.
Data analysis after parallel sequencing is a process that uses combinations of software tools tha... more Data analysis after parallel sequencing is a process that uses combinations of software tools that is often subject to experimentation and on-the-fly substitution, with the necessary file conversion. This article presents a developing system for creating and managing workflows aiding the tasks one encounters after parallel sequences, particularly in the area of metagenomics. The semantics, description language and software implementation aim to allow the creation of flexible, configurable workflows that are suitable for sharing and are easy to manipulate through software or by hand. The execution system design provides user-defined operations and interchangeability between an operation and a workflow. This allows significant extensibility, which can be further complemented with distributed computing and remote management interfaces.
The aim of this paper is to review some methods for text authorship attribution and to discuss th... more The aim of this paper is to review some methods for text authorship attribution and to discuss the development of a software library with tools for automatic authorship attribution. The presentation is focused on an analysis of two groups of tools oriented to: (1) methods for extraction of features and (2) methods for computing the distance between character strings based on data compression algorithms.
The paper analyzes the most signifi cant issues related to the information extraction from docume... more The paper analyzes the most signifi cant issues related to the information extraction from documents and discusses some theoretical models suitable for recognition. It presents, in brief, the main components of a hybrid approach, which combines an application of a domain ontology and deep learning techniques to recognize and classify the document structure effi ciently. The approach is based on the use of a symbolic logic inference model to utilize the knowledge about the document semantics. The document’s segments are referring to the conceptual purpose of the document, which is recognized by applying modern architecture for object detection. An appropriate domain ontology solves the issues related to the semantic completeness of the document’s data. Several experiments have been carried out and the results obtained have been analyzed in terms of the applicability of various modern technologies for the implementation of document understanding system.
The present work presents a survey of the present state of programming languages employing type s... more The present work presents a survey of the present state of programming languages employing type systems with dependent types, i.e., types parameterized by values. The features of the languages are reviewed and compared with an emphasis on the usefulness of the type checker as a tool for asserting correctness statements about programs. Keywords: type systems, dependent types, programming languages, program correctness
The paper presents results from the development of a methodology and corresponding software tools... more The paper presents results from the development of a methodology and corresponding software tools for building an academic repository. The repository was filled up with gaming material. The repository architecture and key features of the search engine are discussed. The emphasis falls on solutions of the large set of problems concerning the development of proper mechanisms for semantics-based search in a digital repository.
The paper presents a methodology for development of tools for semantics oriented search in reposi... more The paper presents a methodology for development of tools for semantics oriented search in repositories of digitized manuscripts. This methodology is based on the application of some Semantic Web technologies to existing manuscript collections that may include: electronic catalogue containing marked-up manuscript descriptions, full texts of manuscripts, digital images of manuscript pages. It is directed to the development of software environments that will be able to deal with complex user queries and answer questions using ontological knowledge and dictionaries of synonyms.
Uploads
Papers by Maria Nisheva