China
Africa
Explore chapters and articles related to this topic
IoT and Big Data Using Intelligence
Published in Vijayalakshmi Saravanan, Alagan Anpalagan, T. Poongodi, Firoz Khan, Securing IoT and Big Data, 2020
Data collected by sensors is fundamentally machine-generated data. Sensors are the managers of the physical computational devices in our IoT ecosystem. They are responsible for the act of sensing and acquiring data from the environment. It can be a machine-to-machine interaction or from an IoT device. This data can be structured, semi-structured, or unstructured. The type of data can be categorized based on data origin. It would be structured if it is from a device, semi-structured if it is from a log file, and unstructured when it is a video or image file. The difference between structured and unstructured data is shown in Figure 3.6. Some examples of machine-generated data can be RFID, GPS output, temperature and other environmental sensing, terrestrial and satellite computer logs, network logs, call records, telemetry which is collected by the government for intelligence purposes, and so on. To define machine-generated data, it is the result of a computer-generated process where human intervention is not involved at all.
Data ownership: Taking stock and mapping the issues
Published in Matthias Dehmer, Frank Emmert-Streib, Frontiers in Data Science, 2017
Florent Thouvenin, Rolf H. Weber, Alfred Fr¨uh
The legal framework is less clear for non-personal data or datasets derived from depersonalized data through Big Data analytics. Such data mostly has an increased value making it more attractive for market players to ask for access. The controller of the data being the original producer is often inclined to retain the data and analyze it in proprietary silos. An increasing amount of machine-generated data is created without direct intervention of an individual by computer processes, applications, or services, or by sensors processing information received from equipment, software, or machinery, whether virtual or real ([7], p. 9). Regardless of such data being stored in-house or in a cloud, third parties are usually denied access. Therefore, a reuse of the data may not occur. As far as trade secrets are concerned, the denial of sharing can be justified. With regard to other data, access might improve its commercialization. Until now, data market places are indeed only slowly emerging ([7], p. 10) evidencing that data exchange is still limited.
Big Data technologies to process spatial and attribute data when designing and operating mine-engineering systems
Published in International Journal of Image and Data Fusion, 2019
Yuri A. Stepanov, Alexander V. Stepanov
Third, to model geodynamic processes of coal-mining enterprise, several static conditions of a model shall be stored, providing each part of the information model with a temporal attribute. It appears possible to forecast a state of geomodel only based on a chronological series produced as a result of modelling. Hence, already high-volume dataset increases in proportion to the number of fixed static states of a model. Thus, many enterprises deal with huge volumes of data, from which the subject area data are needed to be extracted, processed, and analysed to improve efficiency and ensure the security of carrying out mining operations (Ritesh Mehta 2017). The volume is the most obvious but not the only parameter of big data. There arefollow four characteristics inherent in big data: Volume. Machine-generated data substantially exceed the volumes of conventional data. Thus, for instance, one pass of a coal-mining machine enables to generate 10TB of data. If there are 2500 passes per day, the volume of data only from this source will be measured in petabytes. Smart transducers and such complex industrial equipment as processing plants and drilling rigs generate data comparable in volume.Velocity. The flows of data in corporate and social networks are not classified among machine data, but many opinions and comments of users might be helpful to Customer Relationship Management (CRM) systems. Even if a message is limited to 140 symbols, there are 8TB of data generated per day.Diversity. Such traditional formats of data, as relational databases are usually well described, and their structure are relatively stable. The sources of data in big data are most commonly poorly structured, and new types of data are added on a regular basis.Value. Usually, valuable information is hidden in a massive information stream. The task is to retrieve such information, clean it, transform and provide to a user for analysing.