Special Issue on Deep Multi-source Data Analysis (DMDA)
摘要截稿:
全文截稿: 2020-01-31
影响因子: 3.255
期刊难度:
CCF分类: C类
中科院JCR分区:
• 大类 : 计算机科学 - 3区
• 小类 : 计算机:人工智能 - 3区
Overview
Internet makes data acquisition easy and cheap, leading to multi-source data pervasive in the real life. Multi-source data provides enough information that often makes the models can be learned effectively. However, multi-source data is also complex, heterogeneous, and very large in size where inappropriate handling of it will produce ineffective learning models. This inevitably causes the multi-source data analysis a challenging task in many applications. The conversional shallow analysis techniques have been shown to be difficult in dealing with big multi-source data due to their massive volume and multi-source structure, while the most popular deep analysis techniques encounter a lot of limitations (e.g., huge computation power and huge numbers of tuning parameters) in order to make it proficient in the specific domains. Therefore, the study of Deep Multi-source Data Analysis (DMDA) (including novel shallow learning techniques, advanced deep learning techniques, and especially their hybrid) has been a very popular topic in the domain of machine learning and computer vision.
In this special issue, we invite papers to address many challenges of big multi-source data analysis. Specifically, to provide readers of this special issue with state-of-the-art background on the topic,we will invite one survey paper, which will undergo the peer review process. The list of possible topics include, but not limited to:
Multi-source transaction data analysis
Data preprocess of multi-source databases (missing value imputation and feature selection, clustering, and synthesizing/fusion) via shallow learning techniques and deep learning techniques
Distributed/paralleled techniques and sampling techniques for big multi-source databases mining
Transfer learning among multi-source database
Multi-source multimedia data analysis
Representation learning (e.g., deep learning methods, local feature extraction methods, and global feature extraction methods)
Multi-source data analysis tools and applications (e.g. search, storing, ranking, hashing, and retrieval)
Structured/semi-structured multi-source data analysis (e.g., zero-shot learning, one-shot learning, supervised learning, unsupervised learning and semi-supervised learning)
Cross-model data analysis (e.g. search and retrieval) via transfer learning and deep learning
Multi-task data analysis
Similarities/dissimilarities learning from multiple tasks
Regularization strategies in multi-task learning or domain adaptation and transfer learning
Multi-task learning or domain adaptation or transfer learning for big computer vision and multimedia analysis
Large tasks (modals), small sample size learning for multi-task learning, domain adaptation, and transfer learning