Università degli Studi di Napoli "Parthenope"

Teaching schedule

Academic year: 
2021/2022
Belonging course: 
Course of Master's Degree Programme on APPLIED COMPUTER SCIENCE (MACHINE LEARNING AND BIG DATA)
Disciplinary sector: 
INFORMATICS (INF/01)
Language: 
English
Credits: 
6
Year of study: 
1
Teachers: 
Cycle: 
Second semester
Hours of front activity: 
48

Language

Italian

Course description

The course provides the theoretical and practical basis for Signal Processing by using Artificial Intelligence based methodologies. In particular, several Machine Learning and Soft Computing techniques will be considered for audio signal processing.

Knowledge and understanding skills
The student should demonstrate the knowledge and the understanding the basics of the Java and Python programming languages. In particular, it should be able to develop and to analyze methodologies and algorithms for processing and streaming of data from audio signals. The student should have the ability to understand new Java language-based software development methodologies and techniques for audio signal based data analysis.

Application skills
The student should demonstrate his/her acquired knowledge for solving real problems in the signal processing field by using the Java or Python languages. It also should demonstrate the ability to develop and to analyze algorithms complexity and software validation by comparing different signal processing and Artificial Intelligence based processing tools. The student should be able to use and to choice libraries and software systems for the principal application areas.

Judgment autonomy
The student should be able to know how to independently evaluate the effectiveness and efficiency of a software application in real cases.

Communication skills
The student must be able to write a report and a presentation on a software application made by using Java or Python languages and Signal Processing, Machine Learning and Soft Computing data processing tools. The application is developed on topical issues in Multimedia data processing and presents moments of group work. For these reasons the student should demonstrate abilties on these themes.

Ability to learn
The student should be able to autonomously deepening Java-specific topics and applications, such as accessing databases and online software repositories, and other modes available from the network. The student should be able to participate in forums for the continuous updating of knowledge in computer science.

Prerequisites

Basics of Object Oriented Languages

Syllabus

The course provides the theoretical and practical basis for Signal Processing by Artificial Intelligence techniques.

Program

Intelligent Signal Processing

1. Introduction to Multimedia and Signal Processing (8 h)
a. Basics of Multimedia Data Representations
b. Basics of Audio, Image and Video
c. Data Representations

2. Audio and Signal Processing (8 h)
a. Audio Digitization
b. Stochastic Processes
c. Audio Analysis and Transforms
d. Audio Filtering
e. Noisy-Channel Coding
f. Basics of Information Theory and Inference

3. Advanced Signal Processing (8 h)
a. Audio Sound Effects and Synthesis
b. Multimedia Streaming
c. Voice Over IP

4. Soft Computing and Machine Learning for Signal Processing (16 h)
a. Neural Networks based Methodologies
b. Recurrent Neural Networks for time series Prediction
c. Autoencoders for Feature Selection
d. Supervised and Unsupervised Adaptive Learning
e. Signal Filtering, Spectral Estimation, Signal Detection, Signal Reconstruction, Array Signal Processing, System Identification, Signal Compression
f. Principal Component Analysis and Independent Component Analysis
g. Blind Source Separation of Audio
h. Features extraction for Musical Signal Processing
i. Feature Extraction and Data Mining
l. Electroencephalography (EGG) for music perception

5. Soft Computing based Methodologies (8h)
a. Fuzzy Logic for Information Retrieval
b. Neuro-Fuzzy based Systems for Signal Prediction
c. Music Emotion Recognition
d. Sparse Coding and Compressive Sensing
e. Signal Compression
f. Packet Loss in Streaming
g. Signal Reconstruction and Feature Extraction in Bioinformatics

The course aims to provide the basic theoretical and practical foundations for signal processing using Artificial Intelligence techniques. Particular attention will be devoted to the processing of audio signals by Machine Learning and Soft Computing techniques.

Teaching Methods

Teaching is carried out by lectures, seminars by experts in the field and by students themselves. On the e-learning platform there are alternative methods of teaching, learning tests and video lessons.

Textbooks

# e-learning material
# Alan V. Oppenheim, R. W. Schafer; J.R. Buck, Discrete-time signal processing, Upper Saddle River, N.J., Prentice Hall, 1999, ISBN 0-13-754920-2.
# Papoulis, Athanasios; Pillai, S. Unnikrishna (2002). Probability, Random Variables and Stochastic Processes (4th ed.). Boston: McGraw Hill. ISBN 0-07-366011-6.
# MacKay, David J. C., Information Theory, Inference and Learning Algorithms. Cambridge: Cambridge University Press., 2003, ISBN 9780521642989.
# Fundamentals of Multimedia, Z.-N. Li, M. S. Drew, J. Liu, Springer, 2th edition, 2014
# Computer Networking: A Top-Down Approach, J. F. Kurose, K. W. Ross, Pearson, 6 edition, 2013
# T. J. Ross, Fuzzy Logic with Engineering Applications, 4th Edition, 2016
# C. M. Bishop, Pattern Recognition and Machine Learning, 2006

Learning assessment

The student should demonstrated the ability to understand new software development methodologies and techniques based on the Java and / or Python language for the analysis and processing of audio signals. It should be able to elaborate a report and a presentation on a software realized using the Java or Python language on real problems concerning the Multimedia systems. An oral interview allow to evaluate the student's ability to deal with issues related to multimedia systems.
The verification procedure is higlighted precisely in the e-learning platform of the Department of Science and Technology. In summary, the verification procedure consists of an individual project (60% of the overall valutation) and an oral test (40% of the overall valutation).
For the positive outcome of the evaluation the positive outcomes of both tests are necessary (project and oral examination).

More information