Economist highlights the benefits of Big Data and Machine Learning in different areas of knowledge
During the Professional Meeting of the Master in Data Science at INTEC, the economist Lisette Santana explained that Big Data and the techniques inherent to machine learning theory make the search for answers to different issues plausible
SANTO DOMINGO. -It is estimated that humanity produces approximately 2.5 exabytes of information daily, of which 90% has been generated in the last two years. Timely access to this huge amount of data triggered a revolution in various fields of knowledge.
The foregoing was exposed by the economist Lisette Santana, who highlighted that, among other things, large data sets (Big Data) make it plausible to search for answers to different research topics that, probably, would be restricted under a statistical analysis scheme more traditional, as machine learning tools provide a broader analytical framework for decision-making processes.
“Seen in isolation, interest in Big Data topics can be considered as a conjunctural phenomenon, in response to financial crises; however, from another perspective, it reflects a more solid structure with the potential to increase the volume, speed and variety of data at lower costs ”, said the economist.
Santana defined data science as an interdisciplinary field that involves various techniques, processes and systems to derive information and make inference from structured and unstructured data, promoting a scheme under which statistics, applied mathematics, and learning are combined. automatic and other related methods.
"The term Big Data refers to large structured and unstructured data sets that, given their volume and speed of production, cannot be processed using traditional statistical techniques to be useful in a timely manner," he said.
During the virtual conference "Big Data and Machine Learning: indisputable binomial in Data Science", held at the Professional Meeting of the Master of Science in Data of Instituto Tecnológico de Santo Domingo (INTEC), Santana said that under a traditional inference scheme the data is analyzed, a priori assumptions are established about the distribution of information with conservative techniques and approaches, the reduction of inputs is promoted and optimal solutions are sought under certain assumptions .
However, the economist affirmed that under the machine learning scheme, data learning is carried out, there is no rigidity on the information, there is greater freedom in its treatment (more heuristic approach) and it is feasible to manage large data sets, allowing to face more complex problems of learning, dimension, modeling, reasoning and perception.