China
Africa
Explore chapters and articles related to this topic
Fundamentals of Machine Learning
Published in Rashmi Agrawal, Jyotir Moy Chatterjee, Abhishek Kumar, Pramod Singh Rathore, Dac-Nhuong Le, Machine Learning for Healthcare, 2020
In addition to these four dimensions, there are two additional dimensions which are key for operationalizing data. Volatility: this big data dimension refers to the length of time that this data will remain valid and for how long it should be stored.Validity: the validity dimension in big data means that data in use should be correct and accurate. If one wants to use the results for decision making, validity of big data sources and subsequent analysis must be accurate.
Context-Aware Computing
Published in Vivek Kale, Digital Transformation of Enterprise Architecture, 2019
Radio Frequency ID (RFID) tags are widely used in warehouse, retail, library and transport situations, supporting theft detection, stock identification, and tracking and account identity for payment. The tag data depends on the design of the tag, and can range from a simple numeric ID (used in a similar way to a bar-code for stock control), to more complex structured records. The tag computation may range from unauthenticated, unencrypted data exchange on demand, to protocols which require an identifier and provide simple encryption; some protocols may also record tag reading events or cause tag data to be changed. The modification of tag data increases the need for security provision but allows the tag system to function where access to a database indexed by tag data is problematic. The main barrier to invisible tag reading is the need to camouflage long-range antennas; and as possession of tagged objects could reveal quite personal data, their use has been somewhat limited.
Case Studies
Published in Hilary Walton, Security Culture, 2016
To address security awareness and security culture, the organisation put its employees through personal data handling and information security refresher courses to reinforce its key policy messages about data collection, use, disclosure, security, storage and disposal. Key messages were: Patient information should only be used for the purposes for which the information is collected.Log-on to computers using alphanumeric passwords, and change them regularly.Don't ask for customers’ personal data in front of others, and ensure they have privacy when entering PINs.Check signatures and verify that customers are who they say they are.If there has been tampering with terminals or databases, inform management.Keep customer information under lock and key.Shred all confidential waste, including payment card information and photocopies of ID documents.Clean desk tops every night.Only access databases when authorised.Lock systems when not in use.(Credit Institute of Canada 2008)
An evaluation of the retrofit net zero building performances: life cycle energy, emissions and cost
Published in Building Research & Information, 2023
The selected case was the Wayne Aspinall Federal Building and US Courthouse (Aspinall FB-CT), a historic preservation with net-zero energy goals. The decision for this selection is based on three criteria: (i) availability of data, energy use data and cost data, (ii) availability of detailed building construction documents (for quantification of materials), (iii) the highest possible energy efficiency (net zero energy status). It was built in 1918 in Grand Junction, Colorado, and is a three-storey office building with a gross area of 3861 m2. Aspinall FB-CT was built as a rectangular steel and concrete structure with a limestone façade and parapet balustrade (refer to Figure 1). Its original function was as a post office and courthouse; a large extension was added to the east side of the original building in 1938 for office space. The original design and construction did not include an HVAC system, which was added in the 1960s along with an electrical system upgrade. Acoustic ceiling tiles were also added to improve the sound quality as well as conceal the HVAC duct work. The original lighting fixtures were replaced with ceiling-mounted fixtures and wall sconces (Cheng et al., 2014). Since then, there were no major repairs, renovation, or upgrades until 2011. The building was listed on the National Register of Historic Places in 1980 (General Service Adminstration, 2014).
Research on Recommendation System of Online Chinese Learning Resources Based on Multiple Collaborative Filtering Algorithms (RSOCLR)
Published in International Journal of Human–Computer Interaction, 2023
The design of this study refers to the above-mentioned experience and accomplishments, and it has gained some reflective experience from the author team’s practice of the Chinese learning system with analysis and selection functions (Wang & Ge, 2018). Other references come from teaching practice since 2020, particularly communication with learners from various countries and feedback obtained from it. Although previous research has made remarkable achievements in the construction of online Chinese teaching resources and the design of learning resource recommendation system, there is clearly insufficient research on the application of learning resource recommendation methods to the field of online Chinese teaching resources. Therefore, we try to reorganize various data collected through practice, and through discussion and communication with experts in the field of information technology and language education, effectively combine various algorithms based on artificial intelligence (quantitative data) and experts’ suggestions (qualitative data), and use a hybrid algorithm to design an online Chinese learning resource recommendation system.
A framework on task configuration and execution for distributed geographical simulation
Published in International Journal of Digital Earth, 2021
Fengyuan Zhang, Min Chen, Ming Wang, Zihuan Wang, Shuo Zhang, Songshan Yue, Yongning Wen, Guonian Lü
For data transportation optimization, we also use the MD5 of the file to mark the simulation data on the web. Before data transporting, the upload checks the MD5 of the data file and queries the target data exchange nodes if any data share the same MD5 value. If so, users no longer upload the data and use this data address. The data exchange nodes do not provide a data list for data security and have request delays for data checking to avoid cracking. In the data downloading of computing node, in case of data requesting timeout, the data would be downloaded more than twice. If the data downloading is still failed, the running task would be finished and marked with error.