Francesco Lettich, Chiara Pugliese, C. Renso, Fabio Pinelli
The massive use of personal location devices, the Internet of Mobile Things, and Location Based Social Networks, enables the collection of vast amounts of movement data. Such data can be enriched with several semantic dimensions (or aspects), i.e., contextual and heterogeneous information captured in the surrounding environment, leading to the creation of multiple aspect trajectories (MATs). In this work, we present how the MAT-Builder system can be used for the semantic enrichment processing of movement data while being agnostic to aspects and external semantic data sources. This is achieved by integrating MAT-Builder into a methodology which encompasses three design principles and a uniform representation formalism for enriched data based on the Resource Description Framework (RDF) format. An example scenario involving the generation and querying of a dataset of MATs gives a glimpse of the possibilities that our methodology can open up.
{"title":"A general methodology for building multiple aspect trajectories","authors":"Francesco Lettich, Chiara Pugliese, C. Renso, Fabio Pinelli","doi":"10.1145/3555776.3577832","DOIUrl":"https://doi.org/10.1145/3555776.3577832","url":null,"abstract":"The massive use of personal location devices, the Internet of Mobile Things, and Location Based Social Networks, enables the collection of vast amounts of movement data. Such data can be enriched with several semantic dimensions (or aspects), i.e., contextual and heterogeneous information captured in the surrounding environment, leading to the creation of multiple aspect trajectories (MATs). In this work, we present how the MAT-Builder system can be used for the semantic enrichment processing of movement data while being agnostic to aspects and external semantic data sources. This is achieved by integrating MAT-Builder into a methodology which encompasses three design principles and a uniform representation formalism for enriched data based on the Resource Description Framework (RDF) format. An example scenario involving the generation and querying of a dataset of MATs gives a glimpse of the possibilities that our methodology can open up.","PeriodicalId":42971,"journal":{"name":"Applied Computing Review","volume":"6 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81657082","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}
Over the last two decades, Associative Classifiers have shown competitive performance in the task of predicting class labels. Along with the performance in accuracy, associative classifiers produce human-readable predictive rules which is very helpful to understand the decision process of the classifiers. Associative classifiers from early days suffer from the limitation requiring proper threshold value setting which is dataset-specific. Recently some studies eliminated that limitation by producing statistically significant rules. Though recent models showed very competitive performance with state-of-the-art classifiers, their performance is still impacted if the feature vector of the training data is very large. An ensemble model can solve this issue by training each base learner with a subset of the feature vector. In this study, we propose an ensemble model Classification by Frequent Association Rules (CFAR) using associative classifiers as base learners. In our approach, instead of using a classical ensemble and a voting method, we rank the generated rules based on predominance among base learners and select a subset of the rules for predicting class labels. We use 10 datasets from the UCI repository to evaluate the performance of the proposed model. Our ensemble approach CFAR eliminates the limitation of high memory requirement and runtime of recent associative classifiers if training datasets have large feature vectors. Among the datasets we used, along with increasing accuracy in most cases, CFAR removes the noisy rules which enhances the interpretability of the model.
{"title":"Classification by Frequent Association Rules","authors":"Md Rayhan Kabir, Osmar Zaiane","doi":"10.1145/3555776.3577848","DOIUrl":"https://doi.org/10.1145/3555776.3577848","url":null,"abstract":"Over the last two decades, Associative Classifiers have shown competitive performance in the task of predicting class labels. Along with the performance in accuracy, associative classifiers produce human-readable predictive rules which is very helpful to understand the decision process of the classifiers. Associative classifiers from early days suffer from the limitation requiring proper threshold value setting which is dataset-specific. Recently some studies eliminated that limitation by producing statistically significant rules. Though recent models showed very competitive performance with state-of-the-art classifiers, their performance is still impacted if the feature vector of the training data is very large. An ensemble model can solve this issue by training each base learner with a subset of the feature vector. In this study, we propose an ensemble model Classification by Frequent Association Rules (CFAR) using associative classifiers as base learners. In our approach, instead of using a classical ensemble and a voting method, we rank the generated rules based on predominance among base learners and select a subset of the rules for predicting class labels. We use 10 datasets from the UCI repository to evaluate the performance of the proposed model. Our ensemble approach CFAR eliminates the limitation of high memory requirement and runtime of recent associative classifiers if training datasets have large feature vectors. Among the datasets we used, along with increasing accuracy in most cases, CFAR removes the noisy rules which enhances the interpretability of the model.","PeriodicalId":42971,"journal":{"name":"Applied Computing Review","volume":"7 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82466511","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}
F. Donini, A. Marcelletti, A. Morichetta, A. Polini
In inter-organizational contexts, different organizations cooperate exchanging information, to reach specific and shared objectives. The achievement of such interactions raises the need for a trusted communication environment to be used by the participants. This is a particularly relevant challenge when such interactions are specified in a peer-to-peer style, as in the case of Service Choreographies. Indeed, in such situations, the involved participants expect that all the interactions are performed abiding by the agreed specification. To support such a scenario, blockchain technology is gaining interest thanks to its security, trust, and decentralization characteristics. However, technological barriers still limit its adoption in real context due to the costly and time-consuming learning process. For this reason, we propose RESTChain, a general framework relying on blockchain technology enabling in an automatic way the interactions that take place among the participants in a service choreography. Starting from a choreography specification, the framework automatically derives a set of Mediators and a Smart Contract that coordinates the service interactions. In this way, each organization can communicate with the other services through the blockchain in a secure, auditable, and transparent manner.
{"title":"RESTChain: a Blockchain-based Mediator for REST Interactions in Service Choreographies","authors":"F. Donini, A. Marcelletti, A. Morichetta, A. Polini","doi":"10.1145/3555776.3577826","DOIUrl":"https://doi.org/10.1145/3555776.3577826","url":null,"abstract":"In inter-organizational contexts, different organizations cooperate exchanging information, to reach specific and shared objectives. The achievement of such interactions raises the need for a trusted communication environment to be used by the participants. This is a particularly relevant challenge when such interactions are specified in a peer-to-peer style, as in the case of Service Choreographies. Indeed, in such situations, the involved participants expect that all the interactions are performed abiding by the agreed specification. To support such a scenario, blockchain technology is gaining interest thanks to its security, trust, and decentralization characteristics. However, technological barriers still limit its adoption in real context due to the costly and time-consuming learning process. For this reason, we propose RESTChain, a general framework relying on blockchain technology enabling in an automatic way the interactions that take place among the participants in a service choreography. Starting from a choreography specification, the framework automatically derives a set of Mediators and a Smart Contract that coordinates the service interactions. In this way, each organization can communicate with the other services through the blockchain in a secure, auditable, and transparent manner.","PeriodicalId":42971,"journal":{"name":"Applied Computing Review","volume":"35 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74309412","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 progress of computer-aid-diagnosis system for ultrasound breast lesions reaches tremendous success in the past few years. However, conventional deep learning-based strategies in recent developments still have challenges particularly in characterizing tumor domain in ultrasound images due to the heterogeneous and complex variations of lesions along with similar intensity exhibited in target object. To address this, this work proposes a discrete wavelet coefficient-based embeddable branch that allows to additionally propagate geometrical features of tumors in an end-to-end trainable fashion. To be elaborate, such branch priorly enforce the wavelet pooling operation to select a certain coefficient to further collect gradient information of target domain. Further, the current work also investigates two different preprocessing strategies in which the internal and external gradients of lesion areas can be emphasized within the transformation. Thus, we examine the effects of the proposed method based on different preprocessing scenarios. To verify the usefulness, GradCam projection, and the cross-validation demonstrate the connection of the proposed branch encourages the importance of target features, thus boosting the overall discrimination between lesion groups. Lastly, the proposed branch can be easily incorporated with existing deep learning-based architectures.
{"title":"Discrete Wavelet Coefficient-based Embeddable Branch for Ultrasound Breast Masses Classification","authors":"Mingue Song, Yanggon Kim","doi":"10.1145/3555776.3577727","DOIUrl":"https://doi.org/10.1145/3555776.3577727","url":null,"abstract":"The progress of computer-aid-diagnosis system for ultrasound breast lesions reaches tremendous success in the past few years. However, conventional deep learning-based strategies in recent developments still have challenges particularly in characterizing tumor domain in ultrasound images due to the heterogeneous and complex variations of lesions along with similar intensity exhibited in target object. To address this, this work proposes a discrete wavelet coefficient-based embeddable branch that allows to additionally propagate geometrical features of tumors in an end-to-end trainable fashion. To be elaborate, such branch priorly enforce the wavelet pooling operation to select a certain coefficient to further collect gradient information of target domain. Further, the current work also investigates two different preprocessing strategies in which the internal and external gradients of lesion areas can be emphasized within the transformation. Thus, we examine the effects of the proposed method based on different preprocessing scenarios. To verify the usefulness, GradCam projection, and the cross-validation demonstrate the connection of the proposed branch encourages the importance of target features, thus boosting the overall discrimination between lesion groups. Lastly, the proposed branch can be easily incorporated with existing deep learning-based architectures.","PeriodicalId":42971,"journal":{"name":"Applied Computing Review","volume":"56 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74104069","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}
Barbora Buhnova, David Halasz, Danish Iqbal, Hind Bangui
Software systems across various application domains are undergoing a major shift, from static systems of systems to dynamic ecosystems characterized by largely autonomous software agents, engaging in mutual coalitions and partnerships to complete complex collaborative tasks. One of the key challenges facing software engineering along with this shift, is our preparedness to leverage the concept of mutual trust building among the dynamic system components, to support safe collaborations with (possibly malicious or misbehaving) components outside the boundaries of our control. To support safe evolution towards dynamic software ecosystems, this paper examines the current progress in the research on trust in software engineering across various application domains. To this end, it presents a survey of existing work in this area, and suggests the directions in which further research is needed. These directions include the research of social metrics supporting trust assessment, fine-grained quantification of trust-assessment results, and opening the discussion on governance mechanisms responsible for trust-score management and propagation across the integrated software ecosystems.
{"title":"Survey on Trust in Software Engineering for Autonomous Dynamic Ecosystems","authors":"Barbora Buhnova, David Halasz, Danish Iqbal, Hind Bangui","doi":"10.1145/3555776.3577702","DOIUrl":"https://doi.org/10.1145/3555776.3577702","url":null,"abstract":"Software systems across various application domains are undergoing a major shift, from static systems of systems to dynamic ecosystems characterized by largely autonomous software agents, engaging in mutual coalitions and partnerships to complete complex collaborative tasks. One of the key challenges facing software engineering along with this shift, is our preparedness to leverage the concept of mutual trust building among the dynamic system components, to support safe collaborations with (possibly malicious or misbehaving) components outside the boundaries of our control. To support safe evolution towards dynamic software ecosystems, this paper examines the current progress in the research on trust in software engineering across various application domains. To this end, it presents a survey of existing work in this area, and suggests the directions in which further research is needed. These directions include the research of social metrics supporting trust assessment, fine-grained quantification of trust-assessment results, and opening the discussion on governance mechanisms responsible for trust-score management and propagation across the integrated software ecosystems.","PeriodicalId":42971,"journal":{"name":"Applied Computing Review","volume":"127 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75159526","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}
Lilian Berton, Felipe Mitsuishi, Didier Vega Oliveros
In many real-world applications, labeled instances are costly and infeasible to obtain large training sets. This way, learning strategies that do the most with fewer labels are calling attention, such as semi-supervised learning (SSL) and active learning (AL). Active learning allows querying instance to be labeled in the uncertain region and semi-supervised learning classify with a small set of labeled data. We combine both strategies to investigate how AL improves SSL performance, considering both classification results and computational cost. We present experimental results comparing five AL strategies on seven benchmark datasets encompassing synthetic data, handwritten digit and image recognition, and brain-computing interaction tasks. The best single AL strategy was the ranked batch mode, but it has the highest computational cost. On the other hand, using a consensus committee approach leads to the highest results and low-processing footprints.
{"title":"Analysis of active semi-supervised learning","authors":"Lilian Berton, Felipe Mitsuishi, Didier Vega Oliveros","doi":"10.1145/3555776.3577621","DOIUrl":"https://doi.org/10.1145/3555776.3577621","url":null,"abstract":"In many real-world applications, labeled instances are costly and infeasible to obtain large training sets. This way, learning strategies that do the most with fewer labels are calling attention, such as semi-supervised learning (SSL) and active learning (AL). Active learning allows querying instance to be labeled in the uncertain region and semi-supervised learning classify with a small set of labeled data. We combine both strategies to investigate how AL improves SSL performance, considering both classification results and computational cost. We present experimental results comparing five AL strategies on seven benchmark datasets encompassing synthetic data, handwritten digit and image recognition, and brain-computing interaction tasks. The best single AL strategy was the ranked batch mode, but it has the highest computational cost. On the other hand, using a consensus committee approach leads to the highest results and low-processing footprints.","PeriodicalId":42971,"journal":{"name":"Applied Computing Review","volume":"5 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89748429","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}
Because1 the logic bomb performs malicious behaviors only within the branch that triggers the malicious behaviors, if the branch can be easily found, malicious behaviors can be detected efficiently. Existing malicious app analysis tools look for branches that trigger malicious behaviors based on static analysis, so if reflection is used in the app, this branch statement cannot be found properly. Therefore, in this paper, we propose an app execution log-based suspicious conditional statement detection tool that can detect suspicious conditional statements even when reflection is used. The proposed detection tool which is implemented on the android-10.0.0_r47 version of AOSP(Android Open Source Project) can check the branch statement and information about called method while the app is executing, including the method called by reflection. Also, since suspicious conditional statements are detected by checking the method call flow related to branch statements in the execution log, there is no need to examine all branch statements in the app. Experimental results show that the proposed detection tool can detect suspicious conditional statements regardless of the use of reflection.
{"title":"Detecting Suspicious Conditional Statement using App Execution Log","authors":"Sumin Lee, Minho Park, Jiman Hong","doi":"10.1145/3555776.3577722","DOIUrl":"https://doi.org/10.1145/3555776.3577722","url":null,"abstract":"Because1 the logic bomb performs malicious behaviors only within the branch that triggers the malicious behaviors, if the branch can be easily found, malicious behaviors can be detected efficiently. Existing malicious app analysis tools look for branches that trigger malicious behaviors based on static analysis, so if reflection is used in the app, this branch statement cannot be found properly. Therefore, in this paper, we propose an app execution log-based suspicious conditional statement detection tool that can detect suspicious conditional statements even when reflection is used. The proposed detection tool which is implemented on the android-10.0.0_r47 version of AOSP(Android Open Source Project) can check the branch statement and information about called method while the app is executing, including the method called by reflection. Also, since suspicious conditional statements are detected by checking the method call flow related to branch statements in the execution log, there is no need to examine all branch statements in the app. Experimental results show that the proposed detection tool can detect suspicious conditional statements regardless of the use of reflection.","PeriodicalId":42971,"journal":{"name":"Applied Computing Review","volume":"25 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90075158","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}
Thomaz Pereira Da Silva Junior, Everson da Silva Flores, Vagner Santos Da Rosa, F. Borges
This paper presents a comparison of two different types of neural networks when used in the control of a hydraulic actuator. The advantages of using hydraulic actuators are pondered when facing the nonlinearities present in their model, which difficult their control difficult. The state of the art seeks several solutions, mostly in the use of neural networks. In this way, this paper addressed a study regarding the replacement of traditional sigmoidal networks by the use of wavelet networks in the representation of friction on the walls of hydraulic cylinders and reverse valve dynamics. Different architectures are tested and trained using the quickpropagation algorithm. Finally, the efficiency of the networks is compared regarding generalization for friction and reverse dynamics of the valve, as well as their use in a cascade neural control.
{"title":"Machine Learning Applied on Hydraulic Actuator Control","authors":"Thomaz Pereira Da Silva Junior, Everson da Silva Flores, Vagner Santos Da Rosa, F. Borges","doi":"10.1145/3555776.3577695","DOIUrl":"https://doi.org/10.1145/3555776.3577695","url":null,"abstract":"This paper presents a comparison of two different types of neural networks when used in the control of a hydraulic actuator. The advantages of using hydraulic actuators are pondered when facing the nonlinearities present in their model, which difficult their control difficult. The state of the art seeks several solutions, mostly in the use of neural networks. In this way, this paper addressed a study regarding the replacement of traditional sigmoidal networks by the use of wavelet networks in the representation of friction on the walls of hydraulic cylinders and reverse valve dynamics. Different architectures are tested and trained using the quickpropagation algorithm. Finally, the efficiency of the networks is compared regarding generalization for friction and reverse dynamics of the valve, as well as their use in a cascade neural control.","PeriodicalId":42971,"journal":{"name":"Applied Computing Review","volume":"37 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75346363","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 introduce SocioPedia, which is a real-time automatic system for efficiently visualizing and analyzing the variations, characteristics, and evolutions of social knowledge following the change of time. SocioPedia has been developed to provide a full knowledge graph life cycle and combined the temporal information into each processed knowledge. To benefit different classes of users, SocioPedia provides a user-friendly and intuitive environment with different visualization types including static knowledge visualization, timeline knowledge visualization, timeline characteristic visualization, and dynamic timeline visualization.
{"title":"SocioPedia: Visualizing Social Knowledge over Time","authors":"Try My Nguyen, Jason J. Jung","doi":"10.1145/3555776.3577660","DOIUrl":"https://doi.org/10.1145/3555776.3577660","url":null,"abstract":"In this paper, we introduce SocioPedia, which is a real-time automatic system for efficiently visualizing and analyzing the variations, characteristics, and evolutions of social knowledge following the change of time. SocioPedia has been developed to provide a full knowledge graph life cycle and combined the temporal information into each processed knowledge. To benefit different classes of users, SocioPedia provides a user-friendly and intuitive environment with different visualization types including static knowledge visualization, timeline knowledge visualization, timeline characteristic visualization, and dynamic timeline visualization.","PeriodicalId":42971,"journal":{"name":"Applied Computing Review","volume":"1 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78734189","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. Hassanpour, Niels Netten, Tony Busker, Mortaza Shoae Bargh, Sunil Choenni
Machine learning models have been an inevitable tool for analyzing medical images by radiologists. These models provide important information about the contents of these images using extracted radiomic features. However, the dimensionality of the feature space can cause reduction in the accuracy of prediction, a phenomenon known as the curse of dimensionality. In this study we propose a feature selection method using an autoencoder, which incorporates the performance of a classifier within the feature selection process. This is achieved by automatically adjusting a threshold value used for selecting the features fed to the classifier. The contribution of this study is twofold. The first contribution is an improvement to group lasso to include the group size as a cost parameter of the autoencoder. The second contribution is to automate the selection of the threshold value used for eliminating redundant input features. The threshold value in our proposed method is learned during training phase of the proposed model. Our experimental results indicates that the proposed model can successfully converge to appropriate feature selection parameters.
{"title":"Adaptive Feature Selection Using an Autoencoder and Classifier: Applied to a Radiomics Case","authors":"R. Hassanpour, Niels Netten, Tony Busker, Mortaza Shoae Bargh, Sunil Choenni","doi":"10.1145/3555776.3577861","DOIUrl":"https://doi.org/10.1145/3555776.3577861","url":null,"abstract":"Machine learning models have been an inevitable tool for analyzing medical images by radiologists. These models provide important information about the contents of these images using extracted radiomic features. However, the dimensionality of the feature space can cause reduction in the accuracy of prediction, a phenomenon known as the curse of dimensionality. In this study we propose a feature selection method using an autoencoder, which incorporates the performance of a classifier within the feature selection process. This is achieved by automatically adjusting a threshold value used for selecting the features fed to the classifier. The contribution of this study is twofold. The first contribution is an improvement to group lasso to include the group size as a cost parameter of the autoencoder. The second contribution is to automate the selection of the threshold value used for eliminating redundant input features. The threshold value in our proposed method is learned during training phase of the proposed model. Our experimental results indicates that the proposed model can successfully converge to appropriate feature selection parameters.","PeriodicalId":42971,"journal":{"name":"Applied Computing Review","volume":"11 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74668413","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：