China
Africa
Explore chapters and articles related to this topic
Data Mining in IoT
Published in Neeraj Kumar, Aaisha Makkar, Machine Learning in Cognitive IoT, 2020
How can we mine the information from a large amount of data which is produced by IoT devices? Or how can we mine the knowledge from the data as shown in Figure 3.3? We have studied how the data is collected in IoT using sensors. It is quite obvious that using popular data handling schemes such as cloud computing, a large number of knowledgeable patterns are detected. Data extraction is a module that needs special attention. However, the different parameters considered for handling the data are summarized in Figure 3.4 and are described as follows. Data devices: There are numerous IoT devices, which can sense the environment with sensors and produce the data in one form or the other. These devices generate the raw data, which needs to be processed.Data gathering: The data produced by IoT devices are gathered at a particular place by following the procedure of parsing, analyzing, collecting, and merging. This process is responsible for efficient data extraction.Data processing: Data processing is the process of converting raw data into information. Nowadays, open-source solutions are used for this purpose, such as Hadoop.Data services: Data services, such as data storage, data presentation, data extraction, and data binding, are the different data services used.Data tools: Data tools are responsible for data integration, data mining, and data sharing. It starts with the data at network and grows with the advanced tool such as machine learning.
Signal Conditioning and Output Devices
Published in Francis S. Tse, Ivan E. Morse, Measurement and Instrumentation in Engineering, 2018
Since all physical events occur in the time domain, there are time lags in both the analog and digital modes. For analog systems, time lag is manifested in the form of a phase angle in the dynamic response (see Chapter 4). For digital systems, “real-time” data processing is almost instantaneous. Let us not forget that it always takes time to manipulate data. There are time constraints such as in the digital real-time control of machine tools [2].
Project Control System
Published in Adedeji B. Badiru, Project Management, 2019
Processing the data. Data processing is the manipulation of data to generate useful information. Different types of information may be generated from a given data set, depending on how it is processed. The processing method should consider how the information will be used, who will be using it, and what caliber of system response time is desired. If possible, processing controls should be used. This may involve the following steps:
Big data analytics in medical engineering and healthcare: methods, advances and challenges
Published in Journal of Medical Engineering & Technology, 2020
Lidong Wang, Cheryl Ann Alexander
Batch processing, micro-batch processing, and stream processing are three main methods in data processing. Batch processing is used to handle large sets of static data collected over previous time periods while micro-batch processing treats stream data as sequences of smaller data blocks. Stream processing deals with massive sequences of unlimited data that are continuously generated [44]. In-database analytics provides data analysis within a data warehouse, high-speed parallel processing towards big data analytics, and a secure ecosystem for an enterprise with confidential information. This supports preventative healthcare services and improves pharmaceutical management, but in-database analytics is based on the data that are neither in real-time nor current and the analytics is, therefore, probably a static prediction [45]. Big data streaming computing (BDSC) processes high throughput distributed data, offers real-time computation with massively parallel processing and delivers quality healthcare services due to real-time or near real-time decision-making [46].
A 3D body posture analysis framework during merging and lane-changing maneuvers
Published in Journal of Transportation Safety & Security, 2018
A. Kondyli, A. Barmpoutis, V. P. Sisiopiku, L. Zhang, L. Zhao, M. M. Islam, S. S. Patil, S. Rostami Hosuri
As a future step, the proposed algorithm could be further used to predict the intentions of the driver, identify potential hazardous conditions in vehicle's cabin, and warn the driver accordingly through an advanced driver assistance system. For example, by monitoring drivers' body and arm movement through the developed algorithm, it is possible to develop an “average movement profile” for an individual driver, as a function of traffic, geometric, and environmental considerations, and use this profile to identify variations (i.e., due to distraction or other reason) in real time and provide feedback to the driver. For such applications, immediate feedback should be produced, and the computational complexity of the data processing algorithms should be such that real-time execution is feasible.
Machine learning in human resource system of intelligent manufacturing industry
Published in Enterprise Information Systems, 2022
Data preprocessing refers to the corresponding processing operations before the data is fully used. Generally speaking, it includes three levels: data processing, data acquisition and data upload three aspects (Posselt et al. 2016). Data processing refers to the processing and conversion of raw data according to certain requirements and rules, so as to make the data meet the requirements. Finally, data upload refers to the transfer of data from the previous two stages to the data warehouse. The data preprocessing flow is shown in Figure 2.