Pub Date : 2019-07-01DOI: 10.1109/IISA.2019.8900698
Kristen Jaskie, A. Spanias
This paper will address the Positive and Unlabeled learning problem (PU learning) and its importance in the growing field of semi-supervised learning. In most real-world classification applications, well labeled data is expensive or impossible to obtain. We can often label a small subset of data as belonging to the class of interest. It is frequently impractical to manually label all data we are not interested in. We are left with a small set of positive labeled items of interest and a large set of unknown and unlabeled data. Learning a model for this is the PU learning problem.In this paper, we explore several applications for PU learning including examples in biological/medical, business, security, and signal processing. We then survey the literature for new and existing solutions to the PU learning problem.
{"title":"Positive And Unlabeled Learning Algorithms And Applications: A Survey","authors":"Kristen Jaskie, A. Spanias","doi":"10.1109/IISA.2019.8900698","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900698","url":null,"abstract":"This paper will address the Positive and Unlabeled learning problem (PU learning) and its importance in the growing field of semi-supervised learning. In most real-world classification applications, well labeled data is expensive or impossible to obtain. We can often label a small subset of data as belonging to the class of interest. It is frequently impractical to manually label all data we are not interested in. We are left with a small set of positive labeled items of interest and a large set of unknown and unlabeled data. Learning a model for this is the PU learning problem.In this paper, we explore several applications for PU learning including examples in biological/medical, business, security, and signal processing. We then survey the literature for new and existing solutions to the PU learning problem.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132215051","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 : 2019-07-01DOI: 10.1109/IISA.2019.8900683
Alexandros Spournias, Theodoros Skandamis, Eleftherios Pappas, Christos D. Antonopoulos, N. Voros
This paper presents a SLAM technique that does not use odometer information. It is based on HECTOR SLAM method from Technische Universitat of Darmstadt, but using a different hardware from the proposed and finally without the use of IMU device. The method is based on modified settings of the HECTOR SLAM method and manages to optimize the method based on COTS hardware.
{"title":"Enchancing SLAM method for mapping and tracking using a low cost laser scanner","authors":"Alexandros Spournias, Theodoros Skandamis, Eleftherios Pappas, Christos D. Antonopoulos, N. Voros","doi":"10.1109/IISA.2019.8900683","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900683","url":null,"abstract":"This paper presents a SLAM technique that does not use odometer information. It is based on HECTOR SLAM method from Technische Universitat of Darmstadt, but using a different hardware from the proposed and finally without the use of IMU device. The method is based on modified settings of the HECTOR SLAM method and manages to optimize the method based on COTS hardware.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129180031","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 : 2019-07-01DOI: 10.1109/IISA.2019.8900739
Nikolaos Ntaliakouras, Gerasimos Vonitsanos, Andreas Kanavos, Elias Dritsas
Tourism constitutes a vital sector for all countries’ economy and especially for countries like Greece where it holds a significant proportion of the economy. Nowadays, it is crucial for tourism stakeholders to be able to forecast several tourism indicators in order to take appropriate and most profitable decisions. The traditional forecasting models used in tourism are time-series and econometric. In this paper, we propose a methodology which utilizes a data mining technique based on Decision Trees with the aim of providing forecasts for tourism demand taking into account the contribution of explanatory variables. The proposed approach is based on Apache Spark, a robust analytics engine, along with an integrated machine learning library for predicting tourism demand in Greece. The dataset was constructed from publicly available sources and the forecasted (target) variable is the tourist arrivals in Greece for date range 2006 to 2015.
{"title":"An Apache Spark Methodology for Forecasting Tourism Demand in Greece","authors":"Nikolaos Ntaliakouras, Gerasimos Vonitsanos, Andreas Kanavos, Elias Dritsas","doi":"10.1109/IISA.2019.8900739","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900739","url":null,"abstract":"Tourism constitutes a vital sector for all countries’ economy and especially for countries like Greece where it holds a significant proportion of the economy. Nowadays, it is crucial for tourism stakeholders to be able to forecast several tourism indicators in order to take appropriate and most profitable decisions. The traditional forecasting models used in tourism are time-series and econometric. In this paper, we propose a methodology which utilizes a data mining technique based on Decision Trees with the aim of providing forecasts for tourism demand taking into account the contribution of explanatory variables. The proposed approach is based on Apache Spark, a robust analytics engine, along with an integrated machine learning library for predicting tourism demand in Greece. The dataset was constructed from publicly available sources and the forecasted (target) variable is the tourist arrivals in Greece for date range 2006 to 2015.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131244220","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}
The mechanism of random walkers is studied in this paper as a means of disseminating information in resource-constrained networks, such as those encountered in the Internet of Things, a field that is currently emerging in various areas like home networking etc., triggering the creation of new requirements. This leads to an increasing need for effective and drastic solutions. Multiple and replicated random walker mechanisms, which serve as alternative approaches with the ultimate goal of accelerating the coverage of the network, are considered as well. The performance of these mechanisms is experimentally evaluated and compared to past theoretical findings, in a lowcost wireless network deployed on the buildings of a University campus. The obtained experimental results are consistent with previous analytic results, whereas they further indicate the overall time-efficiency in practical applications, showcasing their viability in smart environments.
{"title":"Random Walkers Coverage Experimentation and Evaluation in Low-Cost Wireless Home Networks","authors":"Aikaterini Georgia Alvanou, Konstantinos Skiadopoulos, Konstantinos Giannakis, K. Oikonomou, Georgios Tsoumanis","doi":"10.1109/IISA.2019.8900692","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900692","url":null,"abstract":"The mechanism of random walkers is studied in this paper as a means of disseminating information in resource-constrained networks, such as those encountered in the Internet of Things, a field that is currently emerging in various areas like home networking etc., triggering the creation of new requirements. This leads to an increasing need for effective and drastic solutions. Multiple and replicated random walker mechanisms, which serve as alternative approaches with the ultimate goal of accelerating the coverage of the network, are considered as well. The performance of these mechanisms is experimentally evaluated and compared to past theoretical findings, in a lowcost wireless network deployed on the buildings of a University campus. The obtained experimental results are consistent with previous analytic results, whereas they further indicate the overall time-efficiency in practical applications, showcasing their viability in smart environments.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114411614","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 : 2019-07-01DOI: 10.1109/IISA.2019.8900711
Fragoulis Nikolaos, Ilias Theodorakopoulos, V. Pothos, E. Vassalos
A new, radical CNN dynamic pruning approach is presented in this paper, achieved by a new holistic intervention on both the CNN architecture and the training procedure, which targets to the parsimonious inference by learning to exploit and dynamically remove the redundant capacity of a CNN architecture. Our approach formulates a systematic and data-driven method for developing CNNs that are trained to eventually change size and form in real-time during inference, targeting to the smaller possible computational footprint. Results are provided for the optimal implementation on a few modern, high-end mobile computing platforms indicating a significant speed-up.
{"title":"Dynamic Pruning of CNN networks","authors":"Fragoulis Nikolaos, Ilias Theodorakopoulos, V. Pothos, E. Vassalos","doi":"10.1109/IISA.2019.8900711","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900711","url":null,"abstract":"A new, radical CNN dynamic pruning approach is presented in this paper, achieved by a new holistic intervention on both the CNN architecture and the training procedure, which targets to the parsimonious inference by learning to exploit and dynamically remove the redundant capacity of a CNN architecture. Our approach formulates a systematic and data-driven method for developing CNNs that are trained to eventually change size and form in real-time during inference, targeting to the smaller possible computational footprint. Results are provided for the optimal implementation on a few modern, high-end mobile computing platforms indicating a significant speed-up.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122093488","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 : 2019-07-01DOI: 10.1109/IISA.2019.8900703
Manan Bhagat, M. Alamaniotis, Athanasios Fevgas
The electricity wholesale market is inherently volatile in a deregulated market structure where market participants like power generators and retailors drive the price of electricity. Timely forecasting of the wholesale market prices by market participants has become of utmost importance in order to maximize on profits and minimize on risks. This report presents a hybrid method comprised of an extreme learning machine and a fuzzy inference engine to forecast price intervals using historical wholesale price extreme values (price maximum and minimum), historical load, generation and congestion hours, forecasted temperature and power outage data. This hybrid forecasting method has been tested on RTO Pennsylvania-New Jersey-Maryland (PJM) interconnection for the period July 1st, 2018 to February 8th, 2019, and is compared with individual extreme learning machine and the non-linear autoregressive neural network.
{"title":"Extreme Interval Electricity Price Forecasting of Wholesale Markets Integrating ELM and Fuzzy Inference","authors":"Manan Bhagat, M. Alamaniotis, Athanasios Fevgas","doi":"10.1109/IISA.2019.8900703","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900703","url":null,"abstract":"The electricity wholesale market is inherently volatile in a deregulated market structure where market participants like power generators and retailors drive the price of electricity. Timely forecasting of the wholesale market prices by market participants has become of utmost importance in order to maximize on profits and minimize on risks. This report presents a hybrid method comprised of an extreme learning machine and a fuzzy inference engine to forecast price intervals using historical wholesale price extreme values (price maximum and minimum), historical load, generation and congestion hours, forecasted temperature and power outage data. This hybrid forecasting method has been tested on RTO Pennsylvania-New Jersey-Maryland (PJM) interconnection for the period July 1st, 2018 to February 8th, 2019, and is compared with individual extreme learning machine and the non-linear autoregressive neural network.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"2011 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128929643","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 : 2019-07-01DOI: 10.1109/IISA.2019.8900716
Athanasios Fevgas, Leonidas Akritidis, M. Alamaniotis, P. Tsompanopoulou, Panayiotis Bozanis
The flash based solid state drives have become the storage medium of choice for many applications, replacing traditional HDDs in almost any data center. Their advent has motivated many research efforts in data management. 3DXPoint, a new non-volatile memory with even better specifications, is a breakthrough for storage systems; featuring low latency and high IOPS, 3DXPoint can create new lines of research. Towards this direction, hybrid storage systems combining, both flash and 3DXPoint, seem to be an adequate roadmap, since the cost of 3DXPoint remains high. In this paper, we study the performance of R-tree on both flash and 3DXPoint SSDs through careful experimentation. We also examine a simple yet illuminating approach to develop a hybrid index. The conducted experiments on one real and two synthetic datasets show that spatial indexes can achieve significant performance gains by exploiting 3DXPoint technology. To the best of our knowledge this is the first research effort that considers a hybrid flash/3DXPoint storage for R-tree.
{"title":"A Study of R-tree Performance in Hybrid Flash/3DXPoint Storage","authors":"Athanasios Fevgas, Leonidas Akritidis, M. Alamaniotis, P. Tsompanopoulou, Panayiotis Bozanis","doi":"10.1109/IISA.2019.8900716","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900716","url":null,"abstract":"The flash based solid state drives have become the storage medium of choice for many applications, replacing traditional HDDs in almost any data center. Their advent has motivated many research efforts in data management. 3DXPoint, a new non-volatile memory with even better specifications, is a breakthrough for storage systems; featuring low latency and high IOPS, 3DXPoint can create new lines of research. Towards this direction, hybrid storage systems combining, both flash and 3DXPoint, seem to be an adequate roadmap, since the cost of 3DXPoint remains high. In this paper, we study the performance of R-tree on both flash and 3DXPoint SSDs through careful experimentation. We also examine a simple yet illuminating approach to develop a hybrid index. The conducted experiments on one real and two synthetic datasets show that spatial indexes can achieve significant performance gains by exploiting 3DXPoint technology. To the best of our knowledge this is the first research effort that considers a hybrid flash/3DXPoint storage for R-tree.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131650542","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 : 2019-07-01DOI: 10.1109/IISA.2019.8900676
Konstantinos Papageorgiou, E. Papageorgiou, P. Singh, G. Stamoulis
In this study, we present the functionalities of a new tool for FCMs using credibility weights and OWA-based operators for aggregation tasks. The average aggregation method for weighted interconnections among concepts is the most used method in FCM modeling. The aim of this research work is to (i) propose an alternative aggregation method based on learning OWA operators in aggregating FCM weights, assigned by many experts and/or stakeholders and (ii) to estimate and rank the experts’ credibility using a distance-based method. The applicability and usefulness of the proposed methodology in modeling and decision-making is demonstrated using poverty eradication strategies under DAY-NRLM (Deendayal Antyodaya Yojana-National Rural Livelihoods Mission) of India. The results produced by the proposed learning OWA operators are compared with the known average aggregation method of FCMs. These results imply that the proposed alternative FCM aggregation approach is really challenging when a large number of experts and stakeholders are engaged to design the overall FCM model.
{"title":"A software tool for FCM aggregation employing credibility weights and learning OWA operators","authors":"Konstantinos Papageorgiou, E. Papageorgiou, P. Singh, G. Stamoulis","doi":"10.1109/IISA.2019.8900676","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900676","url":null,"abstract":"In this study, we present the functionalities of a new tool for FCMs using credibility weights and OWA-based operators for aggregation tasks. The average aggregation method for weighted interconnections among concepts is the most used method in FCM modeling. The aim of this research work is to (i) propose an alternative aggregation method based on learning OWA operators in aggregating FCM weights, assigned by many experts and/or stakeholders and (ii) to estimate and rank the experts’ credibility using a distance-based method. The applicability and usefulness of the proposed methodology in modeling and decision-making is demonstrated using poverty eradication strategies under DAY-NRLM (Deendayal Antyodaya Yojana-National Rural Livelihoods Mission) of India. The results produced by the proposed learning OWA operators are compared with the known average aggregation method of FCMs. These results imply that the proposed alternative FCM aggregation approach is really challenging when a large number of experts and stakeholders are engaged to design the overall FCM model.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131673903","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 : 2019-07-01DOI: 10.1109/IISA.2019.8900670
E. Charou, George Felekis, Danai Bournou Stavroulopoulou, Maria Koutsoukou, A. Panagiotopoulou, Yorghos Voutos, E. Bratsolis, Phivos Mylonas, Laurence Likforman-Sulem
Nowadays, governmental programs like ESA’s Copernicus provide freely available data that can be easily utilized for earth observation. In the present work, the problem of detecting agricultural and non-agricultural land cover is addressed. The methodology is based on classification with convolutional neural networks (CNNs) and transfer learning using AlexNet. The study area is located at the Ionian Islands, which include several land cover classes according to Copernicus CORINE Land Cover 2018 (CLC 2018). Furthermore, the dataset consists of natural color images acquired by Sentinel-2A multi-spectral instrument. Experimentation proves that extra addition of training data from foreign grounds, unfamiliar to the Greek data, serves much as a confusing agent regarding network performance.
{"title":"Deep Learning for Agricultural Land Detection in Insular Areas","authors":"E. Charou, George Felekis, Danai Bournou Stavroulopoulou, Maria Koutsoukou, A. Panagiotopoulou, Yorghos Voutos, E. Bratsolis, Phivos Mylonas, Laurence Likforman-Sulem","doi":"10.1109/IISA.2019.8900670","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900670","url":null,"abstract":"Nowadays, governmental programs like ESA’s Copernicus provide freely available data that can be easily utilized for earth observation. In the present work, the problem of detecting agricultural and non-agricultural land cover is addressed. The methodology is based on classification with convolutional neural networks (CNNs) and transfer learning using AlexNet. The study area is located at the Ionian Islands, which include several land cover classes according to Copernicus CORINE Land Cover 2018 (CLC 2018). Furthermore, the dataset consists of natural color images acquired by Sentinel-2A multi-spectral instrument. Experimentation proves that extra addition of training data from foreign grounds, unfamiliar to the Greek data, serves much as a confusing agent regarding network performance.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131051491","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 : 2019-07-01DOI: 10.1109/IISA.2019.8900682
Anastasios Ntourmas, N. Avouris, S. Daskalaki, Y. Dimitriadis
Massive Open Online Courses (MOOCs) offer a wide range of opportunities for learning. Their growing popularity has resulted in a large amount of data being available for learning analytics purposes. A major problem of MOOCs is the overwhelming number of posts in their discussion forums. The forum is a key part of the learning process within a MOOC, so this information overload affects negatively the participants’ learning experience. Automatic classification of the posts can help searching of relevant information for both the learners and teaching assistants. In this study, we address this problem by building two multiclass classification models, using natural language processing techniques, that classify the posts according to a three-category coding scheme. Each model was created with data derived from a MOOC of different subject matter. The main goal was to evaluate each model’s accuracy along with its generalizability to courses of different subject matter. This study contributes to the line of research for automatic classification of forum discussions, ultimately aiming at the development of tools that may assist participants while searching in the forum. Furthermore it provides insights on the main issues that inhibit generalization of classifiers created for a specific subject matter and investigate how their linguistic features relate to this inhibition.
{"title":"Comparative Study of Two Different Mooc Forums Posts Classifiers: Analysis and Generalizability Issues","authors":"Anastasios Ntourmas, N. Avouris, S. Daskalaki, Y. Dimitriadis","doi":"10.1109/IISA.2019.8900682","DOIUrl":"https://doi.org/10.1109/IISA.2019.8900682","url":null,"abstract":"Massive Open Online Courses (MOOCs) offer a wide range of opportunities for learning. Their growing popularity has resulted in a large amount of data being available for learning analytics purposes. A major problem of MOOCs is the overwhelming number of posts in their discussion forums. The forum is a key part of the learning process within a MOOC, so this information overload affects negatively the participants’ learning experience. Automatic classification of the posts can help searching of relevant information for both the learners and teaching assistants. In this study, we address this problem by building two multiclass classification models, using natural language processing techniques, that classify the posts according to a three-category coding scheme. Each model was created with data derived from a MOOC of different subject matter. The main goal was to evaluate each model’s accuracy along with its generalizability to courses of different subject matter. This study contributes to the line of research for automatic classification of forum discussions, ultimately aiming at the development of tools that may assist participants while searching in the forum. Furthermore it provides insights on the main issues that inhibit generalization of classifiers created for a specific subject matter and investigate how their linguistic features relate to this inhibition.","PeriodicalId":371385,"journal":{"name":"2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132749962","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：