Listing 1 - 10 of 10 |
Sort by
|
Choose an application
Anomaly detection refers to the problem of finding rare patterns in data which raise suspicions because they do not comply with an expected behavior. We can consider different kinds of applications like intrusion detection, image processing, system health monitoring and sensor networks. For example, an anomalous pattern coming from sensors on a machine could mean that the machine is ready to break. Most of the current studies on anomaly detection do not consider recent/past events to detect possible new incoming outliers. The use of Long Short-Term Memory (LSTM) networks is then proposed to deal with time dependent data related with anomaly detection problems. The goal of Support Vector Data Description (SVDD) is to describe a realistic domain for the data, excluding superfluous space. The resulting boundary can then be used to detect outliers. In this master thesis, we consider a LSTM-based prediction model for sensor readings coming from a pulp and paper manufacturing machine. Anomalies will then result from too large prediction errors. We compare the SVDD and a discrimination rule based on the assumption of normality for the errors. In the final chapter, we show that for a real world applications the Gaussian distribution for the errors cannot hold and that the need of a non-parametric data descriptions using kernels is real.
Choose an application
As almost all human activities have been moved online due to the pandemic, novel robust and efficient approaches and further research have been in higher demand in the field of computer science and telecommunication. Therefore, this (reprint) book contains 13 high-quality papers presenting advancements in theoretical and practical aspects of computer recognition, pattern recognition, image processing and machine learning (shallow and deep), including, in particular, novel implementations of these techniques in the areas of modern telecommunications and cybersecurity.
Technology: general issues --- History of engineering & technology --- pointer instrumentation --- image processing --- object detection --- K-fold cross-validation --- Faster-RCNN --- vein detection --- digital image processing --- correlation --- displacement measurement --- semantic segmentation --- farmland vacancy segmentation --- strip pooling --- crop growth assessment --- encoder–decoder --- monotone curve --- tangent circle --- adjacent circle --- area of location of the curve --- contour --- fingerprinting --- malware analysis --- malicious network traffic analysis --- HTTP protocol analysis --- pcap file analysis --- malware tracking --- malware identification --- graph theory --- smart meter --- smart metering --- wireless sensor network --- interpolation --- tangent line --- curvature --- error --- ellipse --- B-spline --- dynamic dedicated path protection --- generic Dijkstra algorithm --- elastic optical network --- modulation constraints --- ECG signal --- classification --- PTB-XL --- deep learning --- computer vision --- adversarial attacks --- adversarial defences --- image quality assessment --- stitched images --- panoramic images --- image analysis --- image entropy --- NetFlow --- network intrusion detection --- network behavior analysis --- data quality --- feature selection --- fronthaul --- Xhaul --- DSB-RFoF --- A-RoF --- B5G --- 6G --- DIPP --- optical channel selection --- n/a --- encoder-decoder
Choose an application
This reprint introduces advanced prediction models focused on power load forecasting. Models based on artificial intelligence and more traditional approaches are shown, demonstrating the real possibilities of use to improve prediction in this field. Models of LSTM neural networks, LSTM networks with a SESDA architecture, in even LSTM-CNN are used. On the other hand, multiple seasonal Holt-Winters models with discrete seasonality and the application of the Prophet method to demand forecasting are presented. These models are applied in different circumstances and show highly positive results. This reprint is intended for both researchers related to energy management and those related to forecasting, especially power load.
Research & information: general --- Physics --- Prophet model --- Holt–Winters model --- long-term forecasting --- peak load --- prophet model --- multiple seasonality --- time series --- demand --- load --- forecast --- DIMS --- irregular --- galvanizing --- short-term electrical load forecasting --- machine learning --- deep learning --- statistical analysis --- parameters tuning --- CNN --- LSTM --- short-term load forecast --- Artificial Neural Network --- deep neural network --- recurrent neural network --- attention --- encoder decoder --- online training --- bidirectional long short-term memory --- multi-layer stacked --- neural network --- short-term load forecasting --- power system
Choose an application
This reprint introduces advanced prediction models focused on power load forecasting. Models based on artificial intelligence and more traditional approaches are shown, demonstrating the real possibilities of use to improve prediction in this field. Models of LSTM neural networks, LSTM networks with a SESDA architecture, in even LSTM-CNN are used. On the other hand, multiple seasonal Holt-Winters models with discrete seasonality and the application of the Prophet method to demand forecasting are presented. These models are applied in different circumstances and show highly positive results. This reprint is intended for both researchers related to energy management and those related to forecasting, especially power load.
Prophet model --- Holt–Winters model --- long-term forecasting --- peak load --- prophet model --- multiple seasonality --- time series --- demand --- load --- forecast --- DIMS --- irregular --- galvanizing --- short-term electrical load forecasting --- machine learning --- deep learning --- statistical analysis --- parameters tuning --- CNN --- LSTM --- short-term load forecast --- Artificial Neural Network --- deep neural network --- recurrent neural network --- attention --- encoder decoder --- online training --- bidirectional long short-term memory --- multi-layer stacked --- neural network --- short-term load forecasting --- power system
Choose an application
This reprint introduces advanced prediction models focused on power load forecasting. Models based on artificial intelligence and more traditional approaches are shown, demonstrating the real possibilities of use to improve prediction in this field. Models of LSTM neural networks, LSTM networks with a SESDA architecture, in even LSTM-CNN are used. On the other hand, multiple seasonal Holt-Winters models with discrete seasonality and the application of the Prophet method to demand forecasting are presented. These models are applied in different circumstances and show highly positive results. This reprint is intended for both researchers related to energy management and those related to forecasting, especially power load.
Research & information: general --- Physics --- Prophet model --- Holt–Winters model --- long-term forecasting --- peak load --- prophet model --- multiple seasonality --- time series --- demand --- load --- forecast --- DIMS --- irregular --- galvanizing --- short-term electrical load forecasting --- machine learning --- deep learning --- statistical analysis --- parameters tuning --- CNN --- LSTM --- short-term load forecast --- Artificial Neural Network --- deep neural network --- recurrent neural network --- attention --- encoder decoder --- online training --- bidirectional long short-term memory --- multi-layer stacked --- neural network --- short-term load forecasting --- power system
Choose an application
This book presents recent research results related to various applications of computer vision methods in the widely understood contexts of automation and robotics. As the current progress of image analysis applications may be easily observed in various areas of everyday life, it becomes one of the most essential elements of development of Industry 4.0 solutions. Some of the examples, partially discussed in individual chapters, may be related to the visual navigation of mobile robots and drones, monitoring of industrial production lines, non-destructive evaluation and testing, monitoring of the IoT devices or the 3D printing process and the quality assessment of manufactured objects, video surveillance systems, and decision support in autonomous vehicles.
History of engineering & technology --- machine vision --- defect inspection --- image registration --- feature region --- contour point distribution --- edge gradient direction --- augmented reality --- calibration --- head mounted displays --- optical see-through display --- computer vision --- infrared pedestrian detection --- encoder-decoder --- attention --- convolutional neural network --- deep learning --- domain adaptation --- semantic segmentation --- generative adversarial networks --- convolutional neural networks --- aerial imagery --- image processing --- fertilizers --- distribution --- monitoring --- component association --- part recognition --- feature descriptor --- histogram --- feature subset selection --- industrial objects --- sorting --- k-NN algorithm --- transparent plastic granulate --- recycling --- air nozzles --- additive manufacturing --- 3D prints --- surface quality assessment --- image analysis --- combined metrics --- structural similarity --- video analysis --- visual inspection and diagnostics --- industrial and robotic vision systems
Choose an application
This book presents recent research results related to various applications of computer vision methods in the widely understood contexts of automation and robotics. As the current progress of image analysis applications may be easily observed in various areas of everyday life, it becomes one of the most essential elements of development of Industry 4.0 solutions. Some of the examples, partially discussed in individual chapters, may be related to the visual navigation of mobile robots and drones, monitoring of industrial production lines, non-destructive evaluation and testing, monitoring of the IoT devices or the 3D printing process and the quality assessment of manufactured objects, video surveillance systems, and decision support in autonomous vehicles.
machine vision --- defect inspection --- image registration --- feature region --- contour point distribution --- edge gradient direction --- augmented reality --- calibration --- head mounted displays --- optical see-through display --- computer vision --- infrared pedestrian detection --- encoder-decoder --- attention --- convolutional neural network --- deep learning --- domain adaptation --- semantic segmentation --- generative adversarial networks --- convolutional neural networks --- aerial imagery --- image processing --- fertilizers --- distribution --- monitoring --- component association --- part recognition --- feature descriptor --- histogram --- feature subset selection --- industrial objects --- sorting --- k-NN algorithm --- transparent plastic granulate --- recycling --- air nozzles --- additive manufacturing --- 3D prints --- surface quality assessment --- image analysis --- combined metrics --- structural similarity --- video analysis --- visual inspection and diagnostics --- industrial and robotic vision systems
Choose an application
This book presents recent research results related to various applications of computer vision methods in the widely understood contexts of automation and robotics. As the current progress of image analysis applications may be easily observed in various areas of everyday life, it becomes one of the most essential elements of development of Industry 4.0 solutions. Some of the examples, partially discussed in individual chapters, may be related to the visual navigation of mobile robots and drones, monitoring of industrial production lines, non-destructive evaluation and testing, monitoring of the IoT devices or the 3D printing process and the quality assessment of manufactured objects, video surveillance systems, and decision support in autonomous vehicles.
History of engineering & technology --- machine vision --- defect inspection --- image registration --- feature region --- contour point distribution --- edge gradient direction --- augmented reality --- calibration --- head mounted displays --- optical see-through display --- computer vision --- infrared pedestrian detection --- encoder-decoder --- attention --- convolutional neural network --- deep learning --- domain adaptation --- semantic segmentation --- generative adversarial networks --- convolutional neural networks --- aerial imagery --- image processing --- fertilizers --- distribution --- monitoring --- component association --- part recognition --- feature descriptor --- histogram --- feature subset selection --- industrial objects --- sorting --- k-NN algorithm --- transparent plastic granulate --- recycling --- air nozzles --- additive manufacturing --- 3D prints --- surface quality assessment --- image analysis --- combined metrics --- structural similarity --- video analysis --- visual inspection and diagnostics --- industrial and robotic vision systems
Choose an application
This Special Issue highlights the most recent research being carried out in the NLP field to discuss relative open issues, with a particular focus on both emerging approaches for language learning, understanding, production, and grounding interactively or autonomously from data in cognitive and neural systems, as well as on their potential or real applications in different domains.
tourism big data --- text mining --- NLP --- deep learning --- clinical named entity recognition --- information extraction --- multitask model --- long short-term memory --- conditional random field --- relation extraction --- entity recognition --- long short-term memory network --- multi-turn chatbot --- dialogue context encoding --- WGAN-based response generation --- BERT word embedding --- text summary --- reinforce learning --- FAQ classification --- encoder-decoder neural network --- multi-level word embeddings --- BERT --- bidirectional RNN --- cloze test --- Korean dataset --- machine comprehension --- neural language model --- sentence completion --- primary healthcare --- chief complaint --- virtual medical assistant --- spoken natural language --- disease diagnosis --- medical specialist --- protein–protein interactions --- deep learning (DL) --- convolutional neural networks (CNN) --- bidirectional long short-term memory (bidirectional LSTM) --- dialogue management --- user simulation --- reward shaping --- conversation knowledge --- multi-agent reinforcement learning --- language modeling --- classification --- error probability --- error assessment --- logic error --- neural network --- LSTM --- attention mechanism --- programming education --- neural architecture search --- word ordering --- Korean syntax --- adversarial attack --- adversarial example --- sentiment classification --- dual pointer network --- context-to-entity attention --- text classification --- rule-based --- word embedding --- Doc2vec --- paraphrase identification --- encodings --- R-GCNs --- contextual features --- sentence retrieval --- TF−ISF --- BM25 --- partial match --- sequence similarity --- word to vector --- word embeddings --- antonymy detection --- polarity --- text normalization --- natural language processing --- deep neural networks --- causal encoder --- question classification --- multilingual --- convolutional neural networks --- Natural Language Processing (NLP) --- transfer learning --- open information extraction --- recurrent neural networks --- bilingual translation --- speech-to-text --- LaTeX decompilation --- word representation learning --- word2vec --- sememes --- structural information --- sentiment analysis --- zero-shot learning --- news analysis --- cross-lingual classification --- multilingual transformers --- knowledge base --- commonsense --- sememe prediction --- attention model --- ontologies --- fixing ontologies --- quick fix --- quality metrics --- online social networks --- rumor detection --- Cantonese --- XGA model --- delayed combination --- CNN dictionary --- named entity recognition --- deep learning NER --- bidirectional LSTM CRF --- CoNLL --- OntoNotes --- toxic comments --- neural networks --- n/a --- protein-protein interactions
Choose an application
This Special Issue highlights the most recent research being carried out in the NLP field to discuss relative open issues, with a particular focus on both emerging approaches for language learning, understanding, production, and grounding interactively or autonomously from data in cognitive and neural systems, as well as on their potential or real applications in different domains.
Information technology industries --- Computer science --- tourism big data --- text mining --- NLP --- deep learning --- clinical named entity recognition --- information extraction --- multitask model --- long short-term memory --- conditional random field --- relation extraction --- entity recognition --- long short-term memory network --- multi-turn chatbot --- dialogue context encoding --- WGAN-based response generation --- BERT word embedding --- text summary --- reinforce learning --- FAQ classification --- encoder-decoder neural network --- multi-level word embeddings --- BERT --- bidirectional RNN --- cloze test --- Korean dataset --- machine comprehension --- neural language model --- sentence completion --- primary healthcare --- chief complaint --- virtual medical assistant --- spoken natural language --- disease diagnosis --- medical specialist --- protein-protein interactions --- deep learning (DL) --- convolutional neural networks (CNN) --- bidirectional long short-term memory (bidirectional LSTM) --- dialogue management --- user simulation --- reward shaping --- conversation knowledge --- multi-agent reinforcement learning --- language modeling --- classification --- error probability --- error assessment --- logic error --- neural network --- LSTM --- attention mechanism --- programming education --- neural architecture search --- word ordering --- Korean syntax --- adversarial attack --- adversarial example --- sentiment classification --- dual pointer network --- context-to-entity attention --- text classification --- rule-based --- word embedding --- Doc2vec --- paraphrase identification --- encodings --- R-GCNs --- contextual features --- sentence retrieval --- TF−ISF --- BM25 --- partial match --- sequence similarity --- word to vector --- word embeddings --- antonymy detection --- polarity --- text normalization --- natural language processing --- deep neural networks --- causal encoder --- question classification --- multilingual --- convolutional neural networks --- Natural Language Processing (NLP) --- transfer learning --- open information extraction --- recurrent neural networks --- bilingual translation --- speech-to-text --- LaTeX decompilation --- word representation learning --- word2vec --- sememes --- structural information --- sentiment analysis --- zero-shot learning --- news analysis --- cross-lingual classification --- multilingual transformers --- knowledge base --- commonsense --- sememe prediction --- attention model --- ontologies --- fixing ontologies --- quick fix --- quality metrics --- online social networks --- rumor detection --- Cantonese --- XGA model --- delayed combination --- CNN dictionary --- named entity recognition --- deep learning NER --- bidirectional LSTM CRF --- CoNLL --- OntoNotes --- toxic comments --- neural networks
Listing 1 - 10 of 10 |
Sort by
|