Editorial: Smart Data Management and Applications
- PDF / 151,018 Bytes
- 3 Pages / 595.276 x 790.866 pts Page_size
- 64 Downloads / 222 Views
Editorial: Smart Data Management and Applications Edgar Bisset Alvarez 1 Accepted: 5 September 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Editorial The scientific and technological development achieved by humanity in recent years has led the human being to face a high increase in the amount of data and information available in their surroundings. The large volume of data generated by a diversity of sources, such as mobile applications, sensors, vehicles, buildings, and human beings, also brought about a demand for the correct storage, processing, publication, and verification of the same, to allow each individual to have access to quality information to make the right decisions at certain times in their life. The content produced and disseminated on the web, and especially on online social networks, has posed great challenges for people. At the same time, these social networking environments have allowed for an increase in the sharing, production, use, and consumption of information, opening new perspectives for the evaluation of science, through the so-called social media metrics. Because of this reality, the scientific community has been questioning, are the current data processing techniques sufficient to interpret the content of social media? That is why, researchers from different fields of knowledge such as data science, natural language processing, data engineering, big data, research assessment, network science, sociology of science, and communication are working to face technological, economic, political, and social challenges imposed by the social media metrics of science. This special issue includes eight high-quality articles presented at the First International Conference on Data and Information in Online Environments (DIONE 2020). The first article presents a study of how recurrent neural networks (RNN) are used to make forecasts in the stock market, as well as finding an explanation for the behavior of financial assets and dealing with large volumes of sequences of values. The study points out that most of the studies are * Edgar Bisset Alvarez [email protected] 1
Federal University of Santa Catarina, R. Eng. Agronômico Andrei Cristian Ferreira, s/n - Trindade, Florianópolis, SC 88040-900, Brazil
intended to predict the result of a few stocks. The survey brought together similar market stocks based on their correlation and using the K-averages algorithm, groups of similar stocks were grouped. Subsequently, an RNN was applied to predict possible stock prices. The results showed that the inventory grouping did not influence the network’s efficiency, asthe trend was predicted correctly on average 48% of the time. These results can be of great help to investors, allowing them to buy shares with the same behavior. Besides, the work pointed out that future research includes analyzing the impact of different measures of similarity and how this can influence the process of grouping stocks. The second article proposes an analysis of the privacy principles of personal data
Data Loading...
