The proceedings contain 50 papers. The topics discussed include: drawing on millions of biomedical journal publications to do predictive biology;overview of integrative analysis methods for heterogeneous data;scalable...
ISBN:
(纸本)9781479973033
The proceedings contain 50 papers. The topics discussed include: drawing on millions of biomedical journal publications to do predictive biology;overview of integrative analysis methods for heterogeneous data;scalable graph exploration and visualization: sensemaking challenges and opportunities;distributed real-time knowledge graph serving;serving the readers of scholarly documents: a grand challenge for the introspective digital library;robust network construction against intentional attacks;mind your Ps and Vs: a perspective on the challenges of bigdata management and privacy concerns;exploring concept graphs for biomedical literature mining;MRdatacube: data cube computation using MapReduce;providing QoS through host controlled flash SSD garbage collection and multiple SSDs;promptly pinpointing mobile RFID tags for large-scale Internet-of-things;and fast index construction for distortion-free subsequence matching in time-series databases.
The proceedings contain 93 papers. The topics discussed include: designing and implementing chatbot dialogues for effective knowledge exchange: a methodological approach;combining human-in-the-loop systems and ai fair...
ISBN:
(纸本)9798350370027
The proceedings contain 93 papers. The topics discussed include: designing and implementing chatbot dialogues for effective knowledge exchange: a methodological approach;combining human-in-the-loop systems and ai fairness toolkits to reduce age bias in AI job hiring algorithms;construction of ingredient embedding considering both cooking recipes and their ingredients;learning hierarchy-aware federated graph embedding for link prediction;an empirical study of utility and disclosure risk for tabular data synthesis models: in-depth analysis and interesting findings;unsupervised multi-head attention autoencoder for multivariate time-series anomaly detection;security evaluation of emojis in NLP tasks;a study on the motion sickness reduction technology in vehicle media environment;and emergency landing field identification based on artificial neural networks.
The proceedings contain 95 papers. The topics discussed include: self-supervised learning for climate downscaling;classification of event sequences based on temporal relation features;a machine learning approach to go...
ISBN:
(纸本)9781665475785
The proceedings contain 95 papers. The topics discussed include: self-supervised learning for climate downscaling;classification of event sequences based on temporal relation features;a machine learning approach to government business process re-engineering;pairs trading strategy optimization using proximal policy optimization algorithms;an auditable and efficient prepaid scheme with privacy preservation in smart grids;federated learning with intermediate representation regularization;lazy node-dropping autoencoder;feed-forward design vs. mimic learning for interpretable deep models: a comparative study;inductive graph-based knowledge tracing;fast integration for multiple graphs with Neumann approximation;evaluating mitigation approaches for adversarial attacks in crowdwork;outlier-aware cross-market product recommendation;and edge-cloud collaboration architecture for efficient web-based cognitive services.
The proceedings contain 82 papers. The topics discussed include: comparison of crop image classification model performance according to image augmentation techniques;estimation of compensation scheme functions accordi...
The proceedings contain 82 papers. The topics discussed include: comparison of crop image classification model performance according to image augmentation techniques;estimation of compensation scheme functions according to the level of image data de-identification;trusted and privacy-preserving billing system for smart grids;enhancing gender prediction model performance through automatic individual entity extraction and class balance;an analysis of unsafe responses with magic expressions across large language models;survival sequences: win prediction from a strategy sequence approach;analytical approach for truthfulness detection using multimodal features: a case study with vision and audio data;and exploring the visual perspective taking of vision-and-language model.
The proceedings contain 83 papers. The topics discussed include: transformer-based embedding applied to classify bacterial species using sequencing reads;optimizing performance of real-time bigdata stateful streaming...
The proceedings contain 75 papers. The topics discussed include: STFNet: image classification model based on balanced texture and shape features;multiple 3D LiDARs extrinsic parameter estimation method using plane fea...
ISBN:
(纸本)9781728189246
The proceedings contain 75 papers. The topics discussed include: STFNet: image classification model based on balanced texture and shape features;multiple 3D LiDARs extrinsic parameter estimation method using plane features;a crowd-enabled task execution approach in UAV networks towards fog computing;discovering business problems using problem hypotheses: a goal-oriented and machine learning-based approach;digital healthcare industry and technology trends;preventing enclave malware with intermediate enclaves on semi-honest cloud platforms;pattern-wise embedding system for scalable time-series database;comparison and analysis of embedding methods for patent documents;QUARC: quaternion multi-modal fusion architecture for hate speech classification;and CHNE: context-aware heterogeneous network embedding.
The proceedings contain 121 papers. The topics discussed include: a universal control system for self-driving car towards urban challenges;autonomous driving control using end-to-end deep learning;super-resolution of ...
ISBN:
(纸本)9781728160344
The proceedings contain 121 papers. The topics discussed include: a universal control system for self-driving car towards urban challenges;autonomous driving control using end-to-end deep learning;super-resolution of license plate images via character-based perceptual loss;KISTI vehicle-based urban sensing dataset;deep neural networks on chip - a survey;importance of data distribution on hive-based systems for query performance: an experimental study;de-identification and privacy issues on bigdata transformation;deep learning based response generation using emotion feature extraction;and intellectual priority-based low latency data delivery scheme for multi-interface and multi-channel devices in multi-hop wireless mesh networks.
Quality control is a critical aspect of drug manufacturing. Reducing the number of batches that fail to achieve the required critical quality attributes is a crucial factor for the profitability of the drug manufactur...
详细信息
ISBN:
(纸本)9798350370027;9798350370034
Quality control is a critical aspect of drug manufacturing. Reducing the number of batches that fail to achieve the required critical quality attributes is a crucial factor for the profitability of the drug manufacturer. In the last few decades, several analytical techniques have been developed and employed by manufacturers to track the quality at various stages of production. This paper analyses and validates a strategy that employs Machine Learning (ML) models trained continuously on bigdata to predict the finished drugs' Critical Quality Attributes (CQAs), at the early stages of production. The proposed strategy is easy to integrate into traditional pharmaceutical manufacturing without critical modifications.
As most of the Internet of Things (IoT) applications are event-driven, the emergence of the serverless computing paradigm, which is a natural fit for event-driven applications, is promising to host multi-tenant IoT ap...
详细信息
ISBN:
(纸本)9798350370027;9798350370034
As most of the Internet of Things (IoT) applications are event-driven, the emergence of the serverless computing paradigm, which is a natural fit for event-driven applications, is promising to host multi-tenant IoT applications. Furthermore, the increasing resource capability of low-cost edge and fog devices provides an opportunity to take advantage of resources available and leads to the edge-fog-cloud computing continuum, which can conduct processing across the entire computing continuum. To identify the necessary adaptations for the serverless computing continuum, we integrate the serverless paradigm in each layer of the computing continuum and investigate performance parameters by running serverless workloads using benchmarks.
This paper reports the effect of a novel artificial neural network architecture for industrial anomaly detection using generative adversarial network (GAN)-based data augmentation. We show that GAN-based data augmenta...
详细信息
ISBN:
(纸本)9798350370027;9798350370034
This paper reports the effect of a novel artificial neural network architecture for industrial anomaly detection using generative adversarial network (GAN)-based data augmentation. We show that GAN-based data augmentation enhances the performance of end-to-end electric pole anomaly detection. With the convolutional neural network (CNN) hyperparameter search, our method outperforms vanilla CNN and Cutout augmentation by an average of 2.2%p and 1.6%p, respectively and has an accuracy of over 88% for the test dataset.
暂无评论