P. Wasnik, K. Raja, Ramachandra Raghavendra, C. Busch
Applicability of the face recognition for smartphone-based authentication applications is increasing for different domains such as banking and e-commerce. The unsupervised data capture of face characteristics in biometric applications on smartphones presents the vulnerability to attack the systems using artefact samples. The threat of presentation attacks (a.k.a spoofing attacks) need to be handled to enhance the security of the biometric system. In this work, we present a new approach of using the raw sensor data. We first obtain the residual image corresponding to noise by subtracting the median filtered version of raw data and then computing simple energy value to detect the artefact based presentations. The presented approach uses simple threshold and thereby overcomes the need for learning complex classifiers which are challenging to work on unseen attacks. The proposed method is evaluated using a newly collected database of 390 live presentation attempts of face characteristics and 1530 attack presentations consisting of electronic screen attacks and printed attacks on the iPhone 6S smartphone. Significantly lower average classification error (
{"title":"Presentation Attack Detection in Face Biometric Systems Using Raw Sensor Data from Smartphones","authors":"P. Wasnik, K. Raja, Ramachandra Raghavendra, C. Busch","doi":"10.1109/SITIS.2016.25","DOIUrl":"https://doi.org/10.1109/SITIS.2016.25","url":null,"abstract":"Applicability of the face recognition for smartphone-based authentication applications is increasing for different domains such as banking and e-commerce. The unsupervised data capture of face characteristics in biometric applications on smartphones presents the vulnerability to attack the systems using artefact samples. The threat of presentation attacks (a.k.a spoofing attacks) need to be handled to enhance the security of the biometric system. In this work, we present a new approach of using the raw sensor data. We first obtain the residual image corresponding to noise by subtracting the median filtered version of raw data and then computing simple energy value to detect the artefact based presentations. The presented approach uses simple threshold and thereby overcomes the need for learning complex classifiers which are challenging to work on unseen attacks. The proposed method is evaluated using a newly collected database of 390 live presentation attempts of face characteristics and 1530 attack presentations consisting of electronic screen attacks and printed attacks on the iPhone 6S smartphone. Significantly lower average classification error (","PeriodicalId":403704,"journal":{"name":"2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123522499","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 research project we present in this paper concerns an ontology-based recommender for teacher training and support. An important issue in Computer-Assisted Language Learning education is the complexity of exploiting technology in order to enhance language teaching. The integration of technology into language education is a rather complex achievement, which implies the ability to understand the potential of a tool for language development in different settings. Thus, the aim of the project is to exploit content-based techniques to recommend the most suited instructional design solutions given the mentioned conditions. The tool provides step-by-step recommendations and explanations and is also able to generate scenarios so that it can be used as a valuable training tool. In this context, the paper presents an example of step-by-step recommendation and the semantic modeling of learning tasks, competences and activities. Finally it reports the results of a study aimed to evaluate the correctness of the recommended solutions and explore the potential impact on the end users.
{"title":"A Recommender System as a Support and Training Tool","authors":"Ilaria Torre, Simone Torsani","doi":"10.1109/SITIS.2016.127","DOIUrl":"https://doi.org/10.1109/SITIS.2016.127","url":null,"abstract":"The research project we present in this paper concerns an ontology-based recommender for teacher training and support. An important issue in Computer-Assisted Language Learning education is the complexity of exploiting technology in order to enhance language teaching. The integration of technology into language education is a rather complex achievement, which implies the ability to understand the potential of a tool for language development in different settings. Thus, the aim of the project is to exploit content-based techniques to recommend the most suited instructional design solutions given the mentioned conditions. The tool provides step-by-step recommendations and explanations and is also able to generate scenarios so that it can be used as a valuable training tool. In this context, the paper presents an example of step-by-step recommendation and the semantic modeling of learning tasks, competences and activities. Finally it reports the results of a study aimed to evaluate the correctness of the recommended solutions and explore the potential impact on the end users.","PeriodicalId":403704,"journal":{"name":"2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116843245","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}
Felix Brodkorb, Manuel Kopp, Arjan Kuijper, T. V. Landesberger
Geo-located networks are analyzed in various domains such as supply chain management. When simulating supply chain processes or when testing geo-visualization techniques, synthetic test datasets are needed. However, real world data are difficult to obtain and artificial data are cumbersome to create manually. In this paper, we present an interactive visual tree network generator that not only generates a network, but also attaches geo-locations to its nodes. We designed a modular rulebased system to control the generation process. A user can interactively use rules to parametrize the data generation process. The user can visually explore and adjust results intermediately after each generation iteration.
{"title":"A Modular Rule-Based Visual Interactive Creation of Tree-Shaped Geo-Located Networks","authors":"Felix Brodkorb, Manuel Kopp, Arjan Kuijper, T. V. Landesberger","doi":"10.1109/SITIS.2016.69","DOIUrl":"https://doi.org/10.1109/SITIS.2016.69","url":null,"abstract":"Geo-located networks are analyzed in various domains such as supply chain management. When simulating supply chain processes or when testing geo-visualization techniques, synthetic test datasets are needed. However, real world data are difficult to obtain and artificial data are cumbersome to create manually. In this paper, we present an interactive visual tree network generator that not only generates a network, but also attaches geo-locations to its nodes. We designed a modular rulebased system to control the generation process. A user can interactively use rules to parametrize the data generation process. The user can visually explore and adjust results intermediately after each generation iteration.","PeriodicalId":403704,"journal":{"name":"2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126004486","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}
S. Biondi, Salvatore Monteleone, G. L. Torre, V. Catania
Finding a parking slot is a process very stressful for a person who would like to reach a specific place. It could be considered also expensive since it often causes an increasing of traffic with a consequent increase of pollution. In this paper, we present a smart parking system that helps users to reach a free parking slot, in a small area or a city, using context-aware information to help in the process. The proposed solution does not require an existing infrastructure of sensors spread in parking areas since it exploits the potential of modern smartphones with their capability of recognizing user activity and position. This solution introduces an algorithm which compares routes of different drivers to find relations among them. A mechanism based on Bluetooth Low Energy Advertising (BLE Advertising) is then adopted to detect passengers and reduce the bias that would be introduced by the arrival of multiple users in the same parking area. To test the system we developed an application prototype that enables the gathering of contextual information and displays through a map an approximate distribution of free parking areas close to the user's position. A series of tests were performed to evaluate the system to estimate the advantages of a sensors-less architecture.
{"title":"A Context-Aware Smart Parking System","authors":"S. Biondi, Salvatore Monteleone, G. L. Torre, V. Catania","doi":"10.1109/SITIS.2016.76","DOIUrl":"https://doi.org/10.1109/SITIS.2016.76","url":null,"abstract":"Finding a parking slot is a process very stressful for a person who would like to reach a specific place. It could be considered also expensive since it often causes an increasing of traffic with a consequent increase of pollution. In this paper, we present a smart parking system that helps users to reach a free parking slot, in a small area or a city, using context-aware information to help in the process. The proposed solution does not require an existing infrastructure of sensors spread in parking areas since it exploits the potential of modern smartphones with their capability of recognizing user activity and position. This solution introduces an algorithm which compares routes of different drivers to find relations among them. A mechanism based on Bluetooth Low Energy Advertising (BLE Advertising) is then adopted to detect passengers and reduce the bias that would be introduced by the arrival of multiple users in the same parking area. To test the system we developed an application prototype that enables the gathering of contextual information and displays through a map an approximate distribution of free parking areas close to the user's position. A series of tests were performed to evaluate the system to estimate the advantages of a sensors-less architecture.","PeriodicalId":403704,"journal":{"name":"2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126182519","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}
V. Kamla, J. E. N. Mboula, Jeremie Serge Wouansi Towo, C. T. Djamégni
Computer grids are systems containing heterogeneous, autonomous and geographically distributed nodes. The management of these resources is the works of the meta-scheduler, who allocates work the nodes that are part of a grid, such as clusters, which in turn, have their own local schedulers. In this work we propose a new multi-agent distributed meta-scheduling model. Our model takes, on one hand, benefit from the flexibility of task allocation mode of acquaintances network to reduce the complexity of communication in decision-making, and secondly of the double auction sales to bring a mutual satisfaction between customers and resource providers. The Multi-Attribute Utility Theory (MAUT) is used for a more realistic gain of both. After simulation, through comparative performance analyzes, we show that our model has better contribution in terms of customer and supplier satisfaction than six current state-of-art meta-scheduling algorithms. Other qualitative assets as fault tolerance have to be mentioned.
{"title":"Grid's Acquaintance-Based Multiagent Model of Distributed Meta-Scheduling","authors":"V. Kamla, J. E. N. Mboula, Jeremie Serge Wouansi Towo, C. T. Djamégni","doi":"10.1109/SITIS.2016.55","DOIUrl":"https://doi.org/10.1109/SITIS.2016.55","url":null,"abstract":"Computer grids are systems containing heterogeneous, autonomous and geographically distributed nodes. The management of these resources is the works of the meta-scheduler, who allocates work the nodes that are part of a grid, such as clusters, which in turn, have their own local schedulers. In this work we propose a new multi-agent distributed meta-scheduling model. Our model takes, on one hand, benefit from the flexibility of task allocation mode of acquaintances network to reduce the complexity of communication in decision-making, and secondly of the double auction sales to bring a mutual satisfaction between customers and resource providers. The Multi-Attribute Utility Theory (MAUT) is used for a more realistic gain of both. After simulation, through comparative performance analyzes, we show that our model has better contribution in terms of customer and supplier satisfaction than six current state-of-art meta-scheduling algorithms. Other qualitative assets as fault tolerance have to be mentioned.","PeriodicalId":403704,"journal":{"name":"2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128729049","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}
Content based image retrieval (CBIR) is an application of computer vision which tackles the problem of recovering images in large datasets based on a similarity criterion. The role of CBIR in biomedical field is potentially very important, since every day large volumes of different types of images are produced. An effective and reliable CBIR system can help the decision-making process and support the diagnosis made by the clinician, thanks to the possibility to analyze images similar to the one under test. Many successful CBIR systems use features based on local descriptors for image retrieval. In this work, a Bag-of-Words encoding paradigm based on the Scale Invariant Descriptor (SID) is used to extract robust features from the images. For the evaluation of the proposed technique, three datasets in biomedical field have been used: OASIS, which is an MRI dataset, Emphysema and NEMA, which are instead CT datasets. In order to evaluate the effectiveness and the reliability of the proposed technique also in other application fields, some experiments have been carried out on the ORL facial image dataset, used for biometric applications. The results show that the proposed technique outperforms or is comparable to state-of-art CBIR techniques.
基于内容的图像检索(Content based image retrieval, CBIR)是计算机视觉的一种应用，它解决了基于相似准则的大型数据集中图像的恢复问题。由于每天都会产生大量不同类型的图像，因此CBIR在生物医学领域的作用可能非常重要。由于可以分析与被测图像相似的图像，有效可靠的CBIR系统可以帮助决策过程并支持临床医生的诊断。许多成功的CBIR系统使用基于局部描述符的特征进行图像检索。在这项工作中，使用基于尺度不变描述子(SID)的词袋编码范式从图像中提取鲁棒特征。为了评估所提出的技术，使用了生物医学领域的三个数据集:OASIS (MRI数据集)，Emphysema和NEMA (CT数据集)。为了评估该技术在其他应用领域的有效性和可靠性，在ORL面部图像数据集上进行了一些实验，用于生物识别应用。结果表明，所提出的技术优于或可与最先进的CBIR技术相媲美。
{"title":"Scale Invariant Descriptor for Content Based Image Retrieval in Biomedical Applications","authors":"N. Brancati, Diego Gragnaniello, L. Verdoliva","doi":"10.1109/SITIS.2016.39","DOIUrl":"https://doi.org/10.1109/SITIS.2016.39","url":null,"abstract":"Content based image retrieval (CBIR) is an application of computer vision which tackles the problem of recovering images in large datasets based on a similarity criterion. The role of CBIR in biomedical field is potentially very important, since every day large volumes of different types of images are produced. An effective and reliable CBIR system can help the decision-making process and support the diagnosis made by the clinician, thanks to the possibility to analyze images similar to the one under test. Many successful CBIR systems use features based on local descriptors for image retrieval. In this work, a Bag-of-Words encoding paradigm based on the Scale Invariant Descriptor (SID) is used to extract robust features from the images. For the evaluation of the proposed technique, three datasets in biomedical field have been used: OASIS, which is an MRI dataset, Emphysema and NEMA, which are instead CT datasets. In order to evaluate the effectiveness and the reliability of the proposed technique also in other application fields, some experiments have been carried out on the ORL facial image dataset, used for biometric applications. The results show that the proposed technique outperforms or is comparable to state-of-art CBIR techniques.","PeriodicalId":403704,"journal":{"name":"2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130075804","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 paper presents two systems to recognize five facial expressions (anger, surprise, joy, sadness and neutral) and gives a performance review on them. Both systems are developed on the same facial features extraction process which is histograms of oriented gradients extraction. Vectors of facial features are classified by the systems using the following proposed methods: template matching method based on normalized cross correlation, to find the degree of similarity between inputted images and templates stored in a space of vectors, and supervised learning method of a multi-layer feed-forward neural network. Paper results demonstrate that the adopted methods are efficient, accurate and compete one with other. According to the performance review of these two methods on a three experimental databases (Karolinska Directed Emotional Faces, Cohn-Kanade and Chicago Face Database), normalized cross correlation recognize facial expressions rapidly in high resolutions while neural network is slower but more accurate during classification.
本文提出了两种识别五种面部表情(愤怒、惊讶、喜悦、悲伤和中性)的系统，并对它们进行了性能评估。这两个系统都是在相同的人脸特征提取过程上开发的，即直方图的定向梯度提取。系统对人脸特征向量的分类采用了以下方法:基于归一化互相关的模板匹配方法，用于寻找输入图像与存储在向量空间中的模板之间的相似程度;多层前馈神经网络的监督学习方法。实验结果表明，所采用的方法是有效的、准确的，具有一定的竞争力。通过对这两种方法在Karolinska Directed Emotional Faces、Cohn-Kanade和Chicago Face Database三个实验数据库上的性能评价，归一化相互关系在高分辨率下快速识别面部表情，而神经网络在分类时速度较慢但准确率较高。
{"title":"Performance Review of a Multi-Layer Feed-Forward Neural Network and Normalized Cross Correlation for Facial Expression Identification","authors":"Latifa Greche, N. Es-Sbai, E. Lavendelis","doi":"10.1109/SITIS.2016.43","DOIUrl":"https://doi.org/10.1109/SITIS.2016.43","url":null,"abstract":"The paper presents two systems to recognize five facial expressions (anger, surprise, joy, sadness and neutral) and gives a performance review on them. Both systems are developed on the same facial features extraction process which is histograms of oriented gradients extraction. Vectors of facial features are classified by the systems using the following proposed methods: template matching method based on normalized cross correlation, to find the degree of similarity between inputted images and templates stored in a space of vectors, and supervised learning method of a multi-layer feed-forward neural network. Paper results demonstrate that the adopted methods are efficient, accurate and compete one with other. According to the performance review of these two methods on a three experimental databases (Karolinska Directed Emotional Faces, Cohn-Kanade and Chicago Face Database), normalized cross correlation recognize facial expressions rapidly in high resolutions while neural network is slower but more accurate during classification.","PeriodicalId":403704,"journal":{"name":"2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130142426","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}
In this paper we provide a general process to create the most K-groups based on incomplete profiles and conditional preferences. The proposed approach is shown to be a combination of techniques that each output is the input of the next. It solves three problems: studying the incompletenesses of profiles based on conditional preferences, distances ans similarity measurement, and finding the top credible K-Groups of elements. Our process is efficient it's based on different previous works and tested techniques where each one returns a result that will be adjusted for the next step until the computation of the top K-Group is leaded. We provide a formal semantic of each step, and we describe how each technique provide an outcome that can be exploited according to the general process. Our work is customizable relevant to each approach and algorithm used. We ensure that the top K-Groups formation is reported with no loss in accuracy.
{"title":"Group Formation with Incomplete Profiles","authors":"Zied Ben Othmane, A. Hadjali","doi":"10.1109/SITIS.2016.122","DOIUrl":"https://doi.org/10.1109/SITIS.2016.122","url":null,"abstract":"In this paper we provide a general process to create the most K-groups based on incomplete profiles and conditional preferences. The proposed approach is shown to be a combination of techniques that each output is the input of the next. It solves three problems: studying the incompletenesses of profiles based on conditional preferences, distances ans similarity measurement, and finding the top credible K-Groups of elements. Our process is efficient it's based on different previous works and tested techniques where each one returns a result that will be adjusted for the next step until the computation of the top K-Group is leaded. We provide a formal semantic of each step, and we describe how each technique provide an outcome that can be exploited according to the general process. Our work is customizable relevant to each approach and algorithm used. We ensure that the top K-Groups formation is reported with no loss in accuracy.","PeriodicalId":403704,"journal":{"name":"2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127950001","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}
R. Montella, D. Luccio, P. Troiano, A. Riccio, A. Brizius, Ian T Foster
Accurate prediction of trends in marine pollution is strategic, given the negative effects of low water quality on human marine activities. We describe here the computational and functional performance evaluation of a decision making tool that we developed in the context of an operational workflow for food quality forecast and assessment. Our Water Community Model (WaComM) uses a particle-based Lagrangian approach relying on tridimensional marine dynamics field produced by coupled Eulerian atmosphere and ocean models. WaComM has been developed matching the hierarchical parallelization design requirements and tested in Intel X86_64 and IBM P8 multi core environments and integrated in FACE-IT Galaxy workflow. The predicted pollutant concentration and the amount of pollutants accumulated in the sampled mussels are compared in search of coherent trends to prove the correct model behaviour. In the case study shown in this paper, the predicted Lagrangian tracers, acting as pollutant concentration surrogates, tend to spread rapidly and undergo rapid dilution as expected depending on dominant water column integrated currents.
{"title":"WaComM: A Parallel Water Quality Community Model for Pollutant Transport and Dispersion Operational Predictions","authors":"R. Montella, D. Luccio, P. Troiano, A. Riccio, A. Brizius, Ian T Foster","doi":"10.1109/SITIS.2016.120","DOIUrl":"https://doi.org/10.1109/SITIS.2016.120","url":null,"abstract":"Accurate prediction of trends in marine pollution is strategic, given the negative effects of low water quality on human marine activities. We describe here the computational and functional performance evaluation of a decision making tool that we developed in the context of an operational workflow for food quality forecast and assessment. Our Water Community Model (WaComM) uses a particle-based Lagrangian approach relying on tridimensional marine dynamics field produced by coupled Eulerian atmosphere and ocean models. WaComM has been developed matching the hierarchical parallelization design requirements and tested in Intel X86_64 and IBM P8 multi core environments and integrated in FACE-IT Galaxy workflow. The predicted pollutant concentration and the amount of pollutants accumulated in the sampled mussels are compared in search of coherent trends to prove the correct model behaviour. In the case study shown in this paper, the predicted Lagrangian tracers, acting as pollutant concentration surrogates, tend to spread rapidly and undergo rapid dilution as expected depending on dominant water column integrated currents.","PeriodicalId":403704,"journal":{"name":"2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122707595","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}
Vorapon Luantangsrisuk, Pokpong Songmuang, R. Kongkachandra
An ideal test form should contain questions with different level of difficulties and non-redundant questions. This paper proposed an automated test assembly algorithm to minimize the redundant question in a test form based on Bee algorithm. A neighborhood search in Bee algorithm is improved by using a new technique, called Min-SumDistance (MSD). The MSD is the distance of considered question compared to others in the test form. The sum of question pairs distance indicates to the redundant question in the test form. A question content is represented in two forms as an unigram with TF and TF-IDF scores. The experiments using 200 questions from Information Technology Professional Examination(ITPE). To evaluate the performance of MSD method, we count a number of enemy pairs of the test form and compared to the random method. The experimental results show that our proposed algorithm yields the significant numbers of redundant questions.
{"title":"Automated Test Assembly with Minimum Redundant Questions Based on Bee Algorithm","authors":"Vorapon Luantangsrisuk, Pokpong Songmuang, R. Kongkachandra","doi":"10.1109/SITIS.2016.108","DOIUrl":"https://doi.org/10.1109/SITIS.2016.108","url":null,"abstract":"An ideal test form should contain questions with different level of difficulties and non-redundant questions. This paper proposed an automated test assembly algorithm to minimize the redundant question in a test form based on Bee algorithm. A neighborhood search in Bee algorithm is improved by using a new technique, called Min-SumDistance (MSD). The MSD is the distance of considered question compared to others in the test form. The sum of question pairs distance indicates to the redundant question in the test form. A question content is represented in two forms as an unigram with TF and TF-IDF scores. The experiments using 200 questions from Information Technology Professional Examination(ITPE). To evaluate the performance of MSD method, we count a number of enemy pairs of the test form and compared to the random method. The experimental results show that our proposed algorithm yields the significant numbers of redundant questions.","PeriodicalId":403704,"journal":{"name":"2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133321965","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：