Intelligent systems

Lecture topics:
Intelligence, artificial intelligence (AI) and human-machine interaction: basic philosophical questions about intelligence and AI, the role of AI
Machine learning and data mining, overview of basic algorithms
Data preprocessing, discretization, visualization.
Intelligent data analysis
Basic principles of machine learning (ML), evaluation of learning, combining ML algorithms
Parallel distributed processing and artificial neural networks
Evolutionary computation and genetic algorithms
Basic principles of modelling: learning as modelling, model quality, model evaluation
Statistical modelling: Bayesian reasoning, linear models, regression models, multivariate models, non-parametric models, stochastic processes
Decision support systems: classical decision theory, utility functions, game theory, multi-parameter decision models, uncertainty and risk management, group decision making, quality of decision models
Intelligent agents: overview and state-of-the-art, agent architectures , multiagent systems.
Natural language processing: vector presentation of documents, corpus based methods, information extraction, automatic summarization, text mining.
Reinforcement learning: basic approaches and algorithms, Q learning, TD learning
Heuristic search: minimax principle, alpha-beta pruning, Monte Carlo tree search.

Kononenko, M. Robnik-Šikonja: Inteligentni sistemi, Založba FE in FRI, Ljubljana, 2010.
I. Kononenko, M. Kukar: Machine Learning and Data Mining, Horwood publ., 2007.
S.J. Russell, P. Norvig: Artificial Intelligence: A Modern Approach, 3rd ed. Prentice Hall, 2009.

The goal of the course is the students to become acquainted with the field of intelligent systems, which includes a collection of tools and approaches for solving problems which are difficult or unpractical to tackle with other methods. Students will be able to apply the gained theoretical knowledge on real-world problems from scientific and business environment. The students shall be able to decide which of the presented techniques should be used for a given problem, and to develop a prototype solution.
General competences:
the ability to understand and solve professional challenges in computer and information science,
the ability of professional communication in the native language as well as a foreign language,
the ability to apply acquired knowledge in independent work for solving technical and scientific problems in computer and information science,
familiarity with research methods in the field of computer science.
Subject-specific competences:
using basic machine learning algorithms
preprocessing data for data mining
feature subset selection
evaluation of decision models
using data mining systems
using optimizations packages with evolutionary techniques
text analysis and text mining
using reinforcement learning tools

Knowledge and understanding:
Expertise in several techniques and methods, used for intelligent system modelling. The ability for analysis, synthesis and anticipation of solutions and their consequences for target problems using the scientific methodology.
Application:
The use of the presented methods on target problems from scientific and business environment. The understanding and usage of tools for statistical modelling and data mining.
Reflection:
The recognition and understanding of the meaning of basic mathematical and statistical knowledge, the relation between theory and its application in concrete examples of intelligent modelling and learning. Autonomy, (self) criticalness, (self) reflexivity, aspiration for quality.
Transferable skills:
The transfer of the learned principles to planning of large systems where the principles of intelligent solutions help to improve the usability and the system performance. The ability to receive, select and evaluate new information and a proper interpretation in a context. A self-control and ability to manage limited time when preparing, planning and implementing plans and processes. Team work, writing of reports and articles, public presentations.
Coherent mastering of basic knowledge, gained through mandatory courses, and the ability to combine the knowledge from different fields and to apply it in practice.

Lectures, assignments with written and oral demonstrations and presentations, seminar works and homework. Students from small project teams and autonomously solve assignments based on real-life problems. The teams describe their solutions in written reports and prepare short oral presentations. Written reports and oral presentations are graded.

50% of the final grade is obtained on the basis of on-going work in the laboratory (home-works, quizzes, practical project implementations and presentations).
The other 50% is obtained on the basis of a written exam (this may be complemented by oral exam)
grading: 5 (fail), 6-10 (pass) (according to the Statute of UL)

Igor Kononenko:
KONONENKO, Igor, KUKAR, Matjaž. Machine learning and data mining : introduction to principles and algorithms. Chichester: Horwood Publishing, cop. 2007. XIX, 454 str., ilustr. ISBN 1-904275-21-4. ISBN 978-1-904275-21-3. [COBISS-SI-ID 5961556]
ŠTRUMBELJ, Erik, KONONENKO, Igor. An efficient explanation of individual classifications using game theory. Journal of machine learning research, ISSN 1532-4435. [Print ed.], Jan. 2010, vol. 11, no. [1], str. 1-18, ilustr. [COBISS-SI-ID 7543636]
ROBNIK ŠIKONJA, Marko, KONONENKO, Igor. Theoretical and empirical analysis of ReliefF and RReliefF. Machine learning, ISSN 0885-6125. [Print ed.], 2003, vol. 53, str. 23-69, graf. prikazi. [COBISS-SI-ID 3813460]
KONONENKO, Igor, BRATKO, Ivan. Information-based evaluation criterion for classifier's performance. Machine learning, ISSN 0885-6125. [Print ed.], 1991, vol. 6, no. 1, str. 67-80. [COBISS-SI-ID 7717972]
KONONENKO, Igor. Machine learning for medical diagnosis : history, state of the art and perspective. Artificial intelligence in medicine, ISSN 0933-3657. [Print ed.], 2001, vol. 23, no. 1, str. 89-109. [COBISS-SI-ID 2545236]
Marko Robnik Šikonja:
ROBNIK ŠIKONJA, Marko. Data generators for learning systems based on RBF networks. IEEE transactions on neural networks and learning systems, ISSN 2162-237X. [Print ed.], May 2016, vol. 27, no. 5, str. 926-938, ilustr. , doi: . [COBISS-SI-ID 1536875203]
PIČULIN, Matej, ROBNIK ŠIKONJA, Marko. Handling numeric attributes with ant colony based classifier for medical decision making. Expert systems with applications, ISSN 0957-4174. [Print ed.], Nov. 2014, vol. 41, no. 16, str. 7524-7535, graf. prikazi. [COBISS-SI-ID 10715732]
ROBNIK ŠIKONJA, Marko, VANHOOF, Koen. Evaluation of ordinal attributes at value level. Data mining and knowledge discovery, ISSN 1384-5810, 2007, vol. 14, no. 2, str. [225]-243, ilustr. [COBISS-SI-ID 5801556]
ROBNIK ŠIKONJA, Marko, KONONENKO, Igor. Theoretical and empirical analysis of ReliefF and RReliefF. Machine learning, ISSN 0885-6125. [Print ed.], 2003, vol. 53, str. 23-69, graf. prikazi. [COBISS-SI-ID 3813460]
ROBNIK ŠIKONJA, Marko, KONONENKO, Igor. Explaining classifications for individual instances. IEEE transactions on knowledge and data engineering, ISSN 1041-4347. [Print ed.], May 2008, vol. 20, no. 5, str. 589-600, ilustr. [COBISS-SI-ID 6528340]