Knowledge Based Systems

No precedence is defined for other subjects, nor are recommended basic knowledge specified.

The classes will be taught in a theoretical-practical regime and the teaching methodology will include different pedagogical methods, respectively the expository, demonstrative and project-based learning methods.

The expository method will be used to present the concepts and main contents of the curricular unit. The teacher organizes and orally presents the contents, structuring the reasoning and the result to be obtained. This exhibition will be supported by slides, which will later be made available to students. This exhibition will be complemented with some references made available.

The demonstrative method will be used to exemplify some applications of concepts, namely the application of the different techniques approached for each data mining task. Based on several practical sheets made available, the teacher shares his know-how and demonstrates and helps students in their execution, so that they successfully carry out what is requested there, sometimes on paper, sometimes on a computer, through a tool specific for this purpose.

The project-based learning (PBL) method will be used for the construction of knowledge through a long and continuous work of study, whose purpose is to meet a challenge/problem whose objective is the development of a Knowledge Discovery project in Databases (DCBD), using data from an organization (private or public), open data or creating data through survey.

This course introduces the principles of knowledge-based systems, as well as the theory of knowledge discovery in databases, presenting its concepts, phases, main tasks, techniques and algorithms involved. Knowledge discovery is an area that intersects with the area of artificial intelligence and deals with discovering relationships and patterns in data sets or text, with possible applications in real problems in areas such as marketing, finance, production, telecommunications or fraud detection.

It involves techniques common to machine learning, artificial intelligence and data visualization, such as decision tree induction, artificial neural networks, genetic algorithms, rule induction, fuzzy sets and fuzzy logic, Bayes networks or regression. . At the end, students are challenged to develop a database knowledge discovery project based on a real organization or a theme, or a rule-based knowledge-based system.

The main objectives to be achieved are:

O1 – Know the general architecture and steps in the development of a knowledge-based system

O2 – Know the stages of the knowledge discovery process in databases

O3 – Know and understand the principles of some of the most common techniques in machine learning

O4 – Be able to apply in practice the machine learning techniques discussed

O5- Understand an appropriate methodology to develop a knowledge discovery project in a database

O6 – Develop a database knowledge discovery project based on a real organization or theme

The main competences to be developed are:

C1 – Capacity to elaborate questions that can be answered by a knowledge discovery project in a database

C2 – Capacity to frame a knowledge discovery project in a database and analyze its feasibility

C3 – Ability to propose and create models suitable for specific problems and challenges

C4 – Aptitude to understand existing data from different sources and prepare new data in order to support models

C5 – Ability to analyze the models created according to the perspective of an organization’s business or a theme

C6 – Ability to propose alternative paths, whether related to data, models or tools

Introduction to Knowledge-Based Systems

• Concepts and definitions
• General structure of an KBS
• Process of developing an KBS
• Planning, knowledge representation and inference
• Support tools for building an KBS

Rule Based Systems

• Introduction to rule-based systems
• production rules
• Progressive and regressive reasoning
• Metaknowledge in conflict resolution
• Advantages and disadvantages

Knowledge Discovery in Databases and Data Mining

• Introduction to Database Knowledge
• Stages of the Knowledge Discovery process in a Database
• Data Mining Approaches
• Areas related to Data Mining
• Data Mining Approach

CRISP-DM Methodology

• Business Understanding.
• Data Understanding.
• Data Preparation.
• Modeling.
• Model Evaluation.
• Implementation

Models and Techniques

• Decision tree induction
• artificial neural networks
• genetic algorithms
• rule induction
• fuzzy sets
• Bayes Networks
• Regression

Tools, technologies and application domains

NAO

• Berry, Michael J, & Linoff, Gordon. (1997). Data Mining Techniques: For Marketing, Sales, and Customer Support. New York, USA: John Wiley & Sons, Inc.
• Bigus, Joseph P. (1996). Data Mining with Neural Networks: Solving Business Problems from Application Development to Decision Support. Crawfordsville, Indiana, USA: McGraw-Hill, Inc.
• Chakrabarti, Soumen, Cox, Earl, Frank, Eibe, Güting, Ralf Hartmut, Han, Jiawei, Jiang, Xia, Neapolitan, Richard E. (2008). Data Mining: Know It All. Burlinghton, Massachusetts: Morgan Kaufmann Publishers.
• Fernandes, Anita Maria da Rocha. (2005). Inteligência Artificial: Noções Gerais. Brasil: Visual Books.
• Larose, Daniel T. (2005). Discovering Knowledge in Data: An Introduction to Data Mining. Hoboken, New Jersey: John Wiley & Sons, Inc.
• Santos, Manuel Filipe, & Azevedo, Carla Sousa. (2005). Data Mining: Descoberta de Conhecimento em Bases de Dados. Lisboa: FCA.
• Santos, Maribel Yasmina, & Ramos, Isabel. (2009). Business Intelligence: Tecnologias da informação na gestão de conhecimento (2ª ed.). Lisboa: FCA.
• North, Matthew. (2012). Data mining for the masses: A Global Text Project Book.
• Turban, Efraim, Aronson, Jay, & Liang, Ting-Peng. (2007). Decision Support Systems and Intelligent Systems (7th ed.). New Delhi, India: Prentice Hall. Pearson Education, Inc.
• Witten, Ian H, & Frank, Eibe. (2005). Data Mining: Practical machine learning tools and techniques (2nd ed.). San Francisco: Elsevier.