ادغام کلان داده ها در اینترنت اشیا Big Data Fusion in Internet of Things

Big Data Fusion in Internet of Things The Internet of Things (IoT) brings the real physical world, virtual cyber world and digital world together. Varieties of sensors, such as mobile termainals, cameras, microchips, wearables and even the Internet and socialized human beings, play an important role in IoT. These sensors collect, generate, and preserve a diversity of data with different representations, scales, and densities from various “things”, which offers IoT the ability to measure, infer and understant environments. Integrating things, data and semantic opens opportunities for knowledge discovery, and further makes it possible to provide advanced and intelligent services. Data (information) fusion is an essential and integral part of IoT. Data in IoT characterised by dynamic and heterogeneous leads to inadequacy of simple single-source analysis methods. Data fusion integrates multiple data and knowledge into a consistent, accurate and useful representation, in which the data are fused to high-quality information to provide a reliable decision support. Therefore, it is important to investigate techniques for understanding and resolving issues about data fusion in IoT. However, there are significant barriers to overcome before the potential benefits are fully realized. First, data in IoT comes in large amounts, is a mixture of structured and unstructured information, arrives at speed and can be of uncertain provenance. Many existing solutions become improper due to high computation complexity. Second, data sources in IoT are often of different quality, and with significant differences in coverage, accuracy and timeliness of data, which brings significant challenges to achieve trustworthy data fusion and analytics. Third, managing, extracting and deeply understanding valuable knowledge from multi-modal sources in IoT is the biggest challenge. Big Data fusion in IoT (BDFIoT) calls for advanced techniques that can fuse the knowledge from various data sources organically and efficiently in a machine learning and data mining task. This special issue aims at presenting advanced research results related to big data fusion in Internet of Things. We finally selected 7 papers from a total of 41 submissions after a rigorous review process and panel discussion on their novelty, research significance, technical correctness, evaluation comprehension and presentation. The first paper, titled A Delay-Aware Schedule Method for Distributed Information Fusion with Elastic and Inelastic Traffic by Shen et al., proposes an online scheduling algorithm and its distributed implementation, named Delay-Guaranteed CSMA, in order to guarantee the performance of Distributed Information Fusion (DIF). Both the timing constraints and the historical transmission statistics of sensors are taken into consideration to ensure good delayguaranteed satisfaction and real-time data delivery. The second paper, titled CSF: Crowdsourcing Semantic Fusion for Heterogeneous Media Big Data by Guo et al., proposes a novel solution named Crowdsourcing Semantic Fusion (CSF) that makes use of the collective wisdom of social users and introduces crowdsourcing computing into semantic fusion for overcoming the challenges that manual annotation is inefficient and automatic annotation is inaccurate. This work provides a convenient interface for users to extract semantic information given heterogeneous media documents obtained from the Internet and designs the algorithms to perform the normalization and fusion given heterogeneous semantic objects.