Data Science studies large amounts of data with the aim of extracting useful information for business. It is a multidisciplinary field, intersecting computer science, mathematics and statistics skills. Data Science is the study of data in order to extract detailed information for business. Concretely, it consists of extracting the value of data through scientific methods, mainly statistics, and innovative analysis techniques to transform a piece of data into useful information for a given context and objectives, generating insights and valuable information.
The data scientist analyst helps the company answer questions about the past and future impact of certain business decisions with the aim of answering questions such as:
- what happened
- why it happened
- what can happen
- what can be done with the results of the analysis
In recent times, the sharp increase in affordable computing power and the wide availability of artificial intelligence and machine learning algorithms (previously dominated by university research) have made data processing fast and efficient. This further contribution to data analysis has made data science a booming field, laying the foundations for a near future full of opportunities and capable of increasingly generating coherent and satisfying answers to business questions.
Quantyca has divided its Data Science offering into four macro areas:
Descriptive analysis takes the available data and examines it in order to obtain detailed information about what has happened or is happening about the phenomenon that the data represent. Its main output is characterised by data visualisations such as pie charts, bar graphs, line graphs, tables or dashboards and infographics that help to have a narrative of the insights from the data.
Diagnostic analysis is basically an examination of data to understand why a given event occurred, which may be particularly detailed or in-depth. It is usually done within dashboards or web applications through techniques such as drill-down, data discovery, data mining and correlation, often performed in an orchestrated manner to fully explore the available dataset.
In this area of analysis, historical data is used to make accurate predictions about data patterns that may occur in the future. Quantyca has characterised these analyses with techniques such as machine learning, neural network training, prediction and predictive modelling. In each of these techniques, algorithms are trained to highlight causal connections between one or more chosen variables and the data.
Prescriptive analysis starts from prediction to arrive at further assumptions. It therefore does not merely predict what is likely to happen, but suggests a response or some optimal actions to achieve a certain outcome. It uses graph analysis, simulation, complex event processing, neural networks and machine learning recommendation engines.