The aim is for student to develop a critical statistical sense, which is necessary in the business environment. Traditional statistical methods such as regression models, decision trees or dimensionality reduction are used. More recent methodologies such as Machine Learning and Deep Learning are also taught. Among others, Random Forest, K-Means, Natural Language Processing or Neural Networks are studied.

Data preparation and cleaning (2)

Exploratory data analysis

Pre-processing of data

Noise and outlier detection. 

Processing of missing values 

Treatment of the unbalanced problem

Structured and unstructured data

Evaluation of the distributions of variables

Statistical analysis of data (8)

1. Review of probability and random variables

2. Bernoulli, binomial, Poisson, negative binomial, exponential, normal distribution. 

3. Multidimensional random variables. Joint density and mass. Conditional distributions. Covariance and correlation. Expectation of a random vector, variance and covariance matrix. Independence of random variables.

4. Inferential statistics. sample and population. Central limit theorem. Point and interval estimation. Maximum likelihood estimators. Booststrapping. Student's t-distribution and chi-square. 

5. Hypothesis testing. Power, power ratio, significance level and sample size. Interpretation of p-value. Difference between statistically significant and technically significant.

6. Analysis of variance.

7. Multiple linear regression and logistic regression

8. Factor analysis and PCA 

9. Time series 

Machine Learning (6)

Supervised vs. unsupervised learning. Classification rules, variable and model selection.

K-means

Classification trees

Neural networks

Nearest neighbour

Naïve Bayes

Ensemblelearning.

Natural Language.

Deep Learning (3)

Advanced neural networks.

Markov random fields and Kalman Bucy Filters.

Cellular automation and discrete dynamical systems.

Starting from a basic level, the aim is for student to acquire medium-advanced knowledge and to be able to follow the rest of the subjects, as well as to develop a certain degree of autonomy for the personal learning phase that precedes the master's degree. The programming languages taught are R and Python, as they are the most popular and in demand in the professional field.

This module includes the data extraction phase, where student acquires the skills to work with traditional data instructions , which are currently the most common in companies. It also includes the management of environments properly understood as Big Data, such as Hadoop or Spark, and data collection techniques in social networks such as Twitter or Facebook, web scanning or image collection.

Finally, visualisation techniques will be addressed using the tools most in demand in today's business environment.

Python for data analysis (5)

1. Syntax and Structures of data.

2. Data storage and manipulation

3. Numpy, Pandas, Matplotlib and Seaborn Libraries

4. Projects

instructions data (2)

Static structure of the instructions data.

Introduction

model Entity Relationship

model Relational

Referential Integrity

Relational Algebra

SQL theory

New types of instructions data.

instructions of NoSQL data

subject document: ElasticSearch.

subject graph. Neo4J

subject column: Cassandra.

Visualisation (2)

General visualisation concepts

Storytelling with data

Commercial platforms for visualisation

Data collection techniques (2)

Data Management:

Master Data Management (MDM).

Data mining in environments similar to business (SQL, Hive).

Web scraping.

Images.

Social networking.

Big Data Techniques (3)

Evolution of computer architecture, Computer networks, birth of Big Data.

Parallelisation (MapReduce Paradigm). Hadoop vs Spark.

Big Data Frameworks: Hortonworks, Cloudera, MapR, BDE.

Cloud computing as an enabling technology for Big Data.

Secure access to cloud providers. Notions of network configuration and security.

Amazon Web Services and Big Data tools.

Google's Big Data tools.

The Master aims to provide a solid training in terms of technical knowledge, but also a business vision, so that once the Master is completed, students can act as a bridge between the executive and technical levels of a project. In this way, they will be taught by professionals from leading companies and multinationals, practical and successful cases, seeking to apply concepts acquired in the first two modules. In addition, we have the collaboration of IESE Business School, the Business School of the University of Navarra.

Project management (5)

Project planning: identification, definition and objectives. 

Agile Methodologies

Privacy and transparency. Ethics of artificial intelligence.

Application to the capstone project

Workshops with companies (4)

Presentation of projects and real cases by companies.

Case method.

It plays an important role in the programme. A practical approach is sought that at the same time provides solutions to real problems and projects proposed by companies with which there are collaboration agreements. It can be co-directed by both companies and academics from the University of Navarra, and is an excellent opportunity for students to lead the implementation of projects that have an impact on their professional environment.

Master's Thesis (18)

The TFM will consist of an original work in which the competences acquired during the Master's degree must be put into practice. It can be developed in the framework of a business or institution that proposes a project of collection, cleaning, preparation, advanced analytics and visualisation of the results. 

Ethical aspects of data processing, as well as the economic and social impact of the results, should be highlighted. The student must demonstrate that they know how to plan a project and carry it out in a real working environment, in such a way that they acquire a very practical experience in the field of Data Science and Big Data.