Pub Date : 2019-11-14DOI: 10.1109/IISA.2019.8900770
Gerasimos Vonitsanos, Andreas Kanavos, Alaa Mohasseb, D. Tsolis
Aspect mining constitutes an essential part of delivering concise and, perhaps more importantly, accurately tailored cultural content. With the advent of social media, there is a data abundance so that analytics can be reliably designed for ultimately providing valuable information towards a given product or service. Naturally representing and efficiently processing a large number of opinions can be implemented with the use of streaming technologies. Big data analytics are especially important in the case of cultural content management where reviews and opinions may be analyzed in order to extract meaningful representations. In this paper, a NoSQL database method for aspect mining of a cultural heritage scenario by taking advantage of Apache Spark streaming architecture is presented.
{"title":"A NoSQL Approach for Aspect Mining of Cultural Heritage Streaming Data","authors":"Gerasimos Vonitsanos, Andreas Kanavos, Alaa Mohasseb, D. Tsolis","doi":"10.1109/IISA.2019.8900770","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900770","url":null,"abstract":"Aspect mining constitutes an essential part of delivering concise and, perhaps more importantly, accurately tailored cultural content. With the advent of social media, there is a data abundance so that analytics can be reliably designed for ultimately providing valuable information towards a given product or service. Naturally representing and efficiently processing a large number of opinions can be implemented with the use of streaming technologies. Big data analytics are especially important in the case of cultural content management where reviews and opinions may be analyzed in order to extract meaningful representations. In this paper, a NoSQL database method for aspect mining of a cultural heritage scenario by taking advantage of Apache Spark streaming architecture is presented.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117042580","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900760
Fazal Aman, Azhar Rauf, Rahman Ali, Farkhund Iqbal, A. Khattak
predicting students’ academic performance in advance is of great importance for parents, management of higher education institutions and the student itself. Selection of a right academic program at right time can save time, efforts and resources of both parents and educational institutions. To achieve this goal, an intelligent decision support system (IDSS) is essential to predict students’ performance prior to their admissions in any academic program or getting promoted to the higher classes in an academic program. Scope of this work is to first identify key features, influencing students’ performance, and then develop an accurate predication model for prediction of their performance, prior to taking admission in an intended program or deciding to continue for higher classes and semesters in the same program or to quit the program at this stage. In this study, first, a subjective method is used for identification of academic and socio-economic features to develop the prediction model and then a decision tree-based algorithm, Logistic Model Trees (LMT), is adopted to learn the intrinsic relationship between the identified features and students’ academic grades. The proposed model is trained and tested on a real-world dataset of 1,021 records, collected from examination database of the University of Peshawar. Simulation of the results is performed in Weka 3.8 environment with its default parameters and 10-folds cross validation setting. The proposed system achieved predictive accuracy of 83.48%,which guides parents, management of higher education institutions and students itself to decide whether they should go forward or quit this program at this stage.
{"title":"A Predictive Model for Predicting Students Academic Performance","authors":"Fazal Aman, Azhar Rauf, Rahman Ali, Farkhund Iqbal, A. Khattak","doi":"10.1109/IISA.2019.8900760","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900760","url":null,"abstract":"predicting students’ academic performance in advance is of great importance for parents, management of higher education institutions and the student itself. Selection of a right academic program at right time can save time, efforts and resources of both parents and educational institutions. To achieve this goal, an intelligent decision support system (IDSS) is essential to predict students’ performance prior to their admissions in any academic program or getting promoted to the higher classes in an academic program. Scope of this work is to first identify key features, influencing students’ performance, and then develop an accurate predication model for prediction of their performance, prior to taking admission in an intended program or deciding to continue for higher classes and semesters in the same program or to quit the program at this stage. In this study, first, a subjective method is used for identification of academic and socio-economic features to develop the prediction model and then a decision tree-based algorithm, Logistic Model Trees (LMT), is adopted to learn the intrinsic relationship between the identified features and students’ academic grades. The proposed model is trained and tested on a real-world dataset of 1,021 records, collected from examination database of the University of Peshawar. Simulation of the results is performed in Weka 3.8 environment with its default parameters and 10-folds cross validation setting. The proposed system achieved predictive accuracy of 83.48%,which guides parents, management of higher education institutions and students itself to decide whether they should go forward or quit this program at this stage.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"124 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121208137","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900757
Dimitris Kastaniotis, Dimitrios Tsourounis, Aristotelis Koureleas, Bojidar Peev, C. Theoharatos, S. Fotopoulos
This work focuses on the problem of Lip Reading with Greek words in an unconstrained driving scenario. The goal of Lip Reading (LR) is to understand the spoken work using only visual information, a process also known as Visual Speech Recognition (VSR). This method has several advantages over Speech Recognition, as it can work from a distance and is not affected by other sounds like noise in the environment. In this manner, LR can be considered as an alternative method for speech decoding which can be combined with state-of-the-art speech recognition technologies. The contribution of this work is two-fold. Firstly, a novel dataset with image sequences from Greek words is presented. In total, 10 persons spoke 50 words while they were either driving or simply sitting in the passenger’s seat of a car. The image sequences were recorded with a mobile phone mounted on the windshield of the car. Secondly, the recognition pipeline consists of a Convolutional Neural Network followed by a Long-Short Term Memory Network with a plain attention mechanism. This architecture maps the image sequences to words following an end-to-end learning scheme. Experimental results with various protocols indicate that speaker independent Lip Reading is an extremely challenging problem.
{"title":"Lip Reading in Greek words at unconstrained driving scenario","authors":"Dimitris Kastaniotis, Dimitrios Tsourounis, Aristotelis Koureleas, Bojidar Peev, C. Theoharatos, S. Fotopoulos","doi":"10.1109/IISA.2019.8900757","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900757","url":null,"abstract":"This work focuses on the problem of Lip Reading with Greek words in an unconstrained driving scenario. The goal of Lip Reading (LR) is to understand the spoken work using only visual information, a process also known as Visual Speech Recognition (VSR). This method has several advantages over Speech Recognition, as it can work from a distance and is not affected by other sounds like noise in the environment. In this manner, LR can be considered as an alternative method for speech decoding which can be combined with state-of-the-art speech recognition technologies. The contribution of this work is two-fold. Firstly, a novel dataset with image sequences from Greek words is presented. In total, 10 persons spoke 50 words while they were either driving or simply sitting in the passenger’s seat of a car. The image sequences were recorded with a mobile phone mounted on the windshield of the car. Secondly, the recognition pipeline consists of a Convolutional Neural Network followed by a Long-Short Term Memory Network with a plain attention mechanism. This architecture maps the image sequences to words following an end-to-end learning scheme. Experimental results with various protocols indicate that speaker independent Lip Reading is an extremely challenging problem.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129857315","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900669
Andreas Messalas, Y. Kanellopoulos, C. Makris
The ability to explain in understandable terms, why a machine learning model makes a certain prediction is becoming immensely important, as it ensures trust and transparency in the decision process of the model. Complex models, such as ensemble or deep learning models, are hard to interpret. Various methods have been proposed that deal with this matter. Shapley values provide accurate explanations, as they assign each feature an importance value for a particular prediction. However, the exponential complexity of their calculation is dealt efficiently only in decision tree-based models. Another method is surrogate models, which emulate a black-box model's behavior and provide explanations effortlessly, since they are constructed to be interpretable. Surrogate models are model-agnostic, but they produce only approximate explanations, which cannot always be trusted. We propose a method that combines these two approaches, so that we can take advantage of the model-agnostic part of the surrogate models, as well as the explanatory power of the Shapley values. We introduce a new metric, Topj Similarity, that measures the similitude of two given explanations, produced by Shapley values, in order to evaluate our work. Finally, we recommend ways on how this method could be improved further.
{"title":"Model-Agnostic Interpretability with Shapley Values","authors":"Andreas Messalas, Y. Kanellopoulos, C. Makris","doi":"10.1109/IISA.2019.8900669","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900669","url":null,"abstract":"The ability to explain in understandable terms, why a machine learning model makes a certain prediction is becoming immensely important, as it ensures trust and transparency in the decision process of the model. Complex models, such as ensemble or deep learning models, are hard to interpret. Various methods have been proposed that deal with this matter. Shapley values provide accurate explanations, as they assign each feature an importance value for a particular prediction. However, the exponential complexity of their calculation is dealt efficiently only in decision tree-based models. Another method is surrogate models, which emulate a black-box model's behavior and provide explanations effortlessly, since they are constructed to be interpretable. Surrogate models are model-agnostic, but they produce only approximate explanations, which cannot always be trusted. We propose a method that combines these two approaches, so that we can take advantage of the model-agnostic part of the surrogate models, as well as the explanatory power of the Shapley values. We introduce a new metric, Topj Similarity, that measures the similitude of two given explanations, produced by Shapley values, in order to evaluate our work. Finally, we recommend ways on how this method could be improved further.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130363638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900719
Christos Spatharis, K. Blekas, Alevizos Bastas, T. Kravaris, G. Vouros
In this work we investigate the use of hierarchical collaborative reinforcement learning methods (H-CMARL) for the computation of joint policies to resolve congestion problems in the Air Traffic Management (ATM) domain. In particular, to address cases where the demand of airspace use exceeds capacity, we consider agents representing flights, who need to decide jointly on ground delays at the pre-tactical stage of operations, towards executing their trajectories while adhering to airspace capacity constraints. In doing so, agents collaborate, applying collaborative multi-agent reinforcement learning methods. Specifically, starting from a multiagent Markov Decision Process problem formulation, we introduce a flat and a hierarchical collaborative multiagent reinforcement learning method at two levels (the ground and an abstract one). To quantitatively assess the quality of solutions of the proposed approaches and show the potential of the hierarchical method in resolving the demand-capacity balance problems, we provide experimental results on real-world evaluation cases, where we measure the average delay of flights and the number of flights with delays.
{"title":"Collaborative multiagent reinforcement learning schemes for air traffic management","authors":"Christos Spatharis, K. Blekas, Alevizos Bastas, T. Kravaris, G. Vouros","doi":"10.1109/IISA.2019.8900719","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900719","url":null,"abstract":"In this work we investigate the use of hierarchical collaborative reinforcement learning methods (H-CMARL) for the computation of joint policies to resolve congestion problems in the Air Traffic Management (ATM) domain. In particular, to address cases where the demand of airspace use exceeds capacity, we consider agents representing flights, who need to decide jointly on ground delays at the pre-tactical stage of operations, towards executing their trajectories while adhering to airspace capacity constraints. In doing so, agents collaborate, applying collaborative multi-agent reinforcement learning methods. Specifically, starting from a multiagent Markov Decision Process problem formulation, we introduce a flat and a hierarchical collaborative multiagent reinforcement learning method at two levels (the ground and an abstract one). To quantitatively assess the quality of solutions of the proposed approaches and show the potential of the hierarchical method in resolving the demand-capacity balance problems, we provide experimental results on real-world evaluation cases, where we measure the average delay of flights and the number of flights with delays.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126399950","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900674
K. Tsamis, Andreas Komninos, J. Garofalakis
Social media and microblogs have become an integral part of everyday life. People use microblogs to communicate with each other, express their opinion about a wide range of topics and inform themselves about issues they are interested in. The increasing volume of information generated in microblogs has led to the need of automatically determining the sentiment expressed in microblog comments. Researchers have worked in systematically analyzing microblog comments in order to identify the sentiment expressed in them. Most work in sentiment analysis of microblog comments has been focused on comments written in the English language, whereas fewer efforts have been made in predicting the sentiment of Greek microblog comments. In this paper, we propose a lexicon-based sentiment analysis algorithm for the sentiment classification of both Greek and English microblog comments. The proposed method uses a unified approach for determining the sentiment of comments written in both languages and incorporates techniques that exploit the distinctive features of the language used in microblogs in order to accurately predict the sentiment expressed in microblog comments. Our approach achieves promising results for the sentiment classification of microblog comments into positive, negative or neutral.
{"title":"Leveraging Social Media Linguistic Features for Bilingual Microblog Sentiment Classification","authors":"K. Tsamis, Andreas Komninos, J. Garofalakis","doi":"10.1109/IISA.2019.8900674","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900674","url":null,"abstract":"Social media and microblogs have become an integral part of everyday life. People use microblogs to communicate with each other, express their opinion about a wide range of topics and inform themselves about issues they are interested in. The increasing volume of information generated in microblogs has led to the need of automatically determining the sentiment expressed in microblog comments. Researchers have worked in systematically analyzing microblog comments in order to identify the sentiment expressed in them. Most work in sentiment analysis of microblog comments has been focused on comments written in the English language, whereas fewer efforts have been made in predicting the sentiment of Greek microblog comments. In this paper, we propose a lexicon-based sentiment analysis algorithm for the sentiment classification of both Greek and English microblog comments. The proposed method uses a unified approach for determining the sentiment of comments written in both languages and incorporates techniques that exploit the distinctive features of the language used in microblogs in order to accurately predict the sentiment expressed in microblog comments. Our approach achieves promising results for the sentiment classification of microblog comments into positive, negative or neutral.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132921085","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900687
Spyros Papadimitriou, Eirini Mougiakou, M. Virvou
Intelligent gaming’s appeal in education is growing, thanks to the learning-by-gaming enjoyment factor. Such games, especially artificial intelligence equipped ones, engage in data collection and processing. Both activities fall under the provisions of the new EU data privacy framework, known as GDPR. As such, Authors focus on GDPR principle of personal data processing consent and try to low balance between gaming amusement, educational benefits and regulatory compliance. While doing so, they combine Legal with Computer Sciences with the purpose of proposing applicable solutions with the form of guidelines towards gaming stakeholders in general as well as educational gaming stakeholders in specific.
{"title":"Smart educational games and Consent under the scope of General Data Protection Regulation","authors":"Spyros Papadimitriou, Eirini Mougiakou, M. Virvou","doi":"10.1109/IISA.2019.8900687","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900687","url":null,"abstract":"Intelligent gaming’s appeal in education is growing, thanks to the learning-by-gaming enjoyment factor. Such games, especially artificial intelligence equipped ones, engage in data collection and processing. Both activities fall under the provisions of the new EU data privacy framework, known as GDPR. As such, Authors focus on GDPR principle of personal data processing consent and try to low balance between gaming amusement, educational benefits and regulatory compliance. While doing so, they combine Legal with Computer Sciences with the purpose of proposing applicable solutions with the form of guidelines towards gaming stakeholders in general as well as educational gaming stakeholders in specific.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115678333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900779
C. Troussas, Akrivi Krouska, M. Virvou
Learning management systems have been widely used for managing the learning material and providing assessments to students. However, so far, they fail to offer intelligence and adaptivity in their diagnostic and reasoning mechanisms. In view of the above, this paper presents a novel and smart Learning Management System for tutoring the programming language Java. Our system performs diagnosis of students’ misconceptions based on their syntax and logical programming mistakes. It also takes as input their learning style which is based on the VARK model (Visual-Auditory-Read/Write-Kinesthetic Learner) in order to provide adaptive grain-size instruction to them. “Grain-size” instruction refers to the level of detail of the domain knowledge that a tutoring system provides to students. As such, the adaptive grain-size domain knowledge delivery corresponds to the knowledge levels and needs of the students. The evaluation was conducted using an established framework and student’s t-test and the results of the system show a high level of acceptance of the presented model.
{"title":"Injecting intelligence into learning management systems: The case of adaptive grain-size instruction","authors":"C. Troussas, Akrivi Krouska, M. Virvou","doi":"10.1109/IISA.2019.8900779","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900779","url":null,"abstract":"Learning management systems have been widely used for managing the learning material and providing assessments to students. However, so far, they fail to offer intelligence and adaptivity in their diagnostic and reasoning mechanisms. In view of the above, this paper presents a novel and smart Learning Management System for tutoring the programming language Java. Our system performs diagnosis of students’ misconceptions based on their syntax and logical programming mistakes. It also takes as input their learning style which is based on the VARK model (Visual-Auditory-Read/Write-Kinesthetic Learner) in order to provide adaptive grain-size instruction to them. “Grain-size” instruction refers to the level of detail of the domain knowledge that a tutoring system provides to students. As such, the adaptive grain-size domain knowledge delivery corresponds to the knowledge levels and needs of the students. The evaluation was conducted using an established framework and student’s t-test and the results of the system show a high level of acceptance of the presented model.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124371276","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900722
C. Troussas, Akrivi Krouska, M. Virvou
The proliferation of Internet technologies has rendered education available to a vast majority of people, irrespective of their place, giving birth to e-learning. As such, learners, sharing different characteristics, have access to the learning material. In the light of recent developments, educational software should offer a student-centered learning experience. In view of the above, this paper presents artificial intelligence dynamic clustering of learners’ characteristics for preserving the learning pace of each student. As a testbed of our research, we have designed and implemented an adaptive system for providing individualized tutoring of mathematics to elementary school students. Dynamic clustering takes as input several students’ characteristics, namely pre-existing knowledge, current and previous knowledge level, etc., in order to construct homogeneous student clusters. Through dynamic clustering, the system provides individualized hints to students for improving knowledge acquisition, recommendation for group collaboration, domain knowledge delivery and trophies. The system was evaluated using an established framework and the results show that its incorporated intelligent techniques can offer individualized and adaptive learning while retaining a high level of pedagogical affordance.
{"title":"Adaptive e-learning interactions using dynamic clustering of learners’ characteristics","authors":"C. Troussas, Akrivi Krouska, M. Virvou","doi":"10.1109/IISA.2019.8900722","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900722","url":null,"abstract":"The proliferation of Internet technologies has rendered education available to a vast majority of people, irrespective of their place, giving birth to e-learning. As such, learners, sharing different characteristics, have access to the learning material. In the light of recent developments, educational software should offer a student-centered learning experience. In view of the above, this paper presents artificial intelligence dynamic clustering of learners’ characteristics for preserving the learning pace of each student. As a testbed of our research, we have designed and implemented an adaptive system for providing individualized tutoring of mathematics to elementary school students. Dynamic clustering takes as input several students’ characteristics, namely pre-existing knowledge, current and previous knowledge level, etc., in order to construct homogeneous student clusters. Through dynamic clustering, the system provides individualized hints to students for improving knowledge acquisition, recommendation for group collaboration, domain knowledge delivery and trophies. The system was evaluated using an established framework and the results show that its incorporated intelligent techniques can offer individualized and adaptive learning while retaining a high level of pedagogical affordance.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123660481","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-15DOI: 10.1109/IISA.2019.8900720
D. Spiliotopoulos, C. Vassilakis, Dionisis Margaris, Konstantinos I. Kotis
Natural Language Generators may generate texts that are linguistically enriched. These may result in significantly improved synthetic speech. At the same time, the generators produce pieces of plain text that may span between a single word to a full sentence. Additionally, traditional natural language generators have limited domain coverage, resulting in restricted language analysis of the generated texts. For those cases the enriched input to the speech synthesizer, required for high quality speech synthesis, can be provided by analysing the plain text. This work reports on the method for automatic domain dependent annotation of plain text, through the utilisation of the linguistic information from rich generated text. The synthetic speech from the resulting prosody models is evaluated by human participants showing annotation results for plain text quite on par with the rich generated text. This leads to improved perceived naturalness of the synthesized speech.
{"title":"A Methodology for Generated Text Annotation for High Quality Speech Synthesis","authors":"D. Spiliotopoulos, C. Vassilakis, Dionisis Margaris, Konstantinos I. Kotis","doi":"10.1109/IISA.2019.8900720","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900720","url":null,"abstract":"Natural Language Generators may generate texts that are linguistically enriched. These may result in significantly improved synthetic speech. At the same time, the generators produce pieces of plain text that may span between a single word to a full sentence. Additionally, traditional natural language generators have limited domain coverage, resulting in restricted language analysis of the generated texts. For those cases the enriched input to the speech synthesizer, required for high quality speech synthesis, can be provided by analysing the plain text. This work reports on the method for automatic domain dependent annotation of plain text, through the utilisation of the linguistic information from rich generated text. The synthetic speech from the resulting prosody models is evaluated by human participants showing annotation results for plain text quite on par with the rich generated text. This leads to improved perceived naturalness of the synthesized speech.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"1999 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128259967","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：