Algorithms
We use a wide spectrum of algorithms to model the problems to be solved. Some examples:
Neural Networks
Model based on imitating brain neurons. A set of units, called artificial neurons, are defined, connected to each other and the information goes through the neural network producing output values.
Cluster Analysis
Statistical models of grouping observations from the detection of similar behavior patterns
Genetic Algorithms
Modeling resulting from iterative processes from a set of restrictive conditions to the solution
SVM (Support-Vector Machine)
Supervised learning models used for classification and regression analysis
PCA (Principal Component Analysis)
Technique used in exploratory data analysis and predictive models.
Gradient Boosting
Model for regression and classification that produces a predictive model usually in the form of decision trees.
Techniques
We use a wide variety of data analysis techniques. Among them we can highlight:
Data Visualization
Dashboards for reporting and data exploration
Feature Engineering
Establishment of new composite variables from raw data to be used as input for the models.
Data Augmentation
Data increase from variations on an existing data set
Data Enrichment
Data expansion, connection between the organization's own databases and open data or external providers databases.
Data Wrangling
Data hygiene and standardization for subsequent statistical treatment.
Subscribe to our newsletter
