Towards Smart Data Technologies for Big Data Analytics
Currently the publicly available datasets for Big Data Ana-lytics are of different qualities, and obtaining the expected behavior from the Machine Learning algorithms is crucial. Furthermore, since working with a huge amount of data is usually a time-demanding task, tohave high quality data is requi...
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Objeto de conferencia |
| Lenguaje: | Inglés |
| Publicado: |
2020
|
| Materias: | |
| Acceso en línea: | http://sedici.unlp.edu.ar/handle/10915/104775 |
| Aporte de: |
| Sumario: | Currently the publicly available datasets for Big Data Ana-lytics are of different qualities, and obtaining the expected behavior from the Machine Learning algorithms is crucial. Furthermore, since working with a huge amount of data is usually a time-demanding task, tohave high quality data is required. Smart Data refers to the process of transforming Big Data into clean and reliable data, and this can be accomplished by converting them, reducing unnecessary volume of data or applying some preprocessing techniques with the aim of improve their quality, and still to obtain trustworthy results. We present those properties that affect the quality of data. Also, the available proposals to analyze the quality of huge amount of data and to cope with low quality datasets in an scalable way, are commented. Furthermore, the need for a methodology towards Smart Data is highlighted. |
|---|