Courses

Courses of the 1st semester

Machine Learning (DWS101)

Course Content (Syllabus)
Introduction, Regression, Decision Trees, Rule Learning, Instance – Based Learning, Bayesian Learning, Learning with Genetic Algorithms, Model Evaluation, Clustering, Association Rules, Feature Selection and Discretization, Ensemble Methods, Reinforcement Learning, Text Mining, Machine Learning Software (Python).

Learning Outcomes
Students become familiar with the main concepts as well as with specific data analysis and machine learning techniques and become also familiar with many applications of it. They also acquire skills in reviewing scientific papers, giving scientific presentations and working in practice with data using various algorithms and relevant machine learning software.

Technologies for Big Data Analytics (DWS102)

Course Content (Syllabus)
– Introduction to Big Data Management and Analytics
– Hadoop: basic and advanved topics
– The Hadoop ecosystem: HDFS, Hbase, Pig, Hive
– NoSQL databases
– Theoretical issues in MapReduce
– The Scala programming language
– The Spark platform: basic and advanced issues
– Streaming, SQL, Machine Learning, GraphΧ: the basic libraries
– Data exploration using SparkR
– Algorithm design in Spark
– Graph databases
– Other systems: Giraph, GraphLab, Hama, BlinkDB

Learning Outcomes
1. Students will get important knowledge in big data management and analytics
2. They will work in teams
3. They will be more confident by presenting their work in class
4. They will get in contact with modern big data analytics techniques with a lot of applications in Industry.

Semantic Web (DWS103)

Course Content (Syllabus)
Introduction and General vision of the Semantic Web (SW). SW Architecture. Technologies and Languages of the SW. XML (Description, DTD, XML Schema, Namespaces, XPath, XSLT, XML tools). RDF (Description, XML syntax, RDF Schema, RDF/RDFS Semantics, Querying RDF/RDFS with SPARQL, Linked Open Data, RDF tools). OWL (Introduction to ontologies and OWL, Description and syntax, OWL flavors, Examples, OWL in OWL, Future extensions, OWL tools). OWL2 Presentation. Ontology Engineering (Ontology creation, Reusing ontologies, Semi – automatic methods). SW Applications. Linked Open Data. Logic and Inferencing (SWRL, OWL2 RL, RIF, RuleML).

Learning Outcomes

Knowledge: Familiarization with principles and technologies for representing and reasoning about data, metadata, and knowledge in the Semantic Web, Familiarization with Ontology Engineering techniques, Training on XML editors/processors and Ontology editors.

Skills: Developing metadata vocabularies and ontologies, Representation of data, metadata, knowledge and ontologies using the following languages: XML, DTD, XML Schema, XSLT, XPATH, RDF, RDF Schema, SPARQL, OWL and SWRL.

Natural Language Processing (DWS104)

Course Content (Syllabus)
Text processing, N-gram language models, naive Bayes and sentiment classification, vector semantics and embeddings, hidden Markov models and conditional random fields for sequence classification (part-of-speech tagging, named entity recognition), neural networks and neural language models, deep learning architectures for sequence processing, applications (relation extraction, keyphrase extraction, summarization, question answering, chatbots and dialog systems)

Social Network Analysis (DWS105)

Course Content (Syllabus)
Centrality measures, network models (random) and characteristics (small world phenomena, power law distribution on various networks characteristics), mechanisms for network generation (preferential attachment), network robustness. Online social data sources, community detection, sentiment analytics basics, information diffusion, influence detection, fraud detection.

Learning Outcomes
1. Identification of social networks and related analysis methods
2. Construction and analysis tools for data that are not in network form (text, image, etc.)
3. Learning of important properties and process characteristics on networks like community detection, sentiment analysis, information diffusion, network robustness, etc.

Managing and Mining Complex Networks (New)

Course Content (Syllabus)
The aim of the course is to deepen in issues of processing and analysis of complex networks. More specifically, the topics covered are: graph data management and analysis, visualization, reachability queries, fast property testing, graph representations learning using scalable techniaues, discovering dense signatures, finding triangles in graphs, graph streams and applications, knowledge extraction from specific types of graphs (e.g. probabilistic, multi-layered, hidden), graph kernels, graph spectral analysis, graph sketches, applications, current trends. The lectures are accompanied by teamwork that should be prepared during the course.

Algorithmic Game Theory (New)

Course Content (Syllabus)
The aim of the course is to study algorithmic techniques in order to analyze and design algorithms in environments where selfish users compete to share resources. More specifically, the topics covered are: introduction to game theory, load balancing games, routing games, cost sharing games, introduction to mechanism design, mechanisms without money, combinatorial auctions, profit maximization, current trends in algorithmic game theory. The lectures are accompanied by teamwork that should be prepared during the course.

Statistical Data Analysis

Course Content (Syllabus)
Statistical methods in data analysis using the R language: Descriptive statistics and graphical representation of data. Discrete and continuous distributions, random numbers generators and distributions. Statistical inference with parametric and non-parametric methods. Hypothesis tests and confidence intervals by resampling methods. Regression models (linear regression and generalized models for continuous, binary, categorical, count dependent variables and mixed independent variables). Non-parametric regression. Multivariate analysis: Factor analysis, cluster analysis and correlation analysis.

Courses of the 2nd semester

Web Mining (DWS201)

Course Content (Syllabus)
• Introduction to the basic concepts of information and data management within World Wide Web.
• Data types on the web and representation.
• World Wide Web Structure and Graph Model.
• World Wide Web Performance Metrics and Information Sharing Techniques.
• Information aggregation techniques in Social Networks.
• Sentiment analysis and behavioral analytics on the Web.
• Recommendations in Social Media and New Collaborative Web Environments.

Learning Outcomes
1. Identify data structures and models World Wide Web
2. Construct and analyze Web data models and understand data relevance metrics i
3. Learning important Web properties, attributes and algorithms to find and recognize common patterns, behaviors, and Web clusters

Distributed Data Processing (DWS202)

Course Content (Syllabus)
Basic concepts of Distributed Databases (DDB), Distributed Database Systems, DDB Architecture, Autonomy and Heterogeneity Issues, Design of DDBs and Data Allocation, Distributed query processing and optimization, Distributed Database Transactions, reliability and data replication, Parallel databases and relationship with DDBs, Grid data management, Adaptive query processing, Map-Reduce techniques, Modern systems, NoSQL DBMSs and current research issues.

Learning Outcomes
Training on distributed data management topics. Acquisition of practical and theoretical knowledge of through the attendance of lectures and the implementation of projects.

Mining from Massive Datasets (DWS203)

Course Content (Syllabus)
– Introduction to Big Data Management and Analytics
– Hadoop: basic and advanved topics
– The Hadoop ecosystem: HDFS, Hbase, Pig, Hive
– NoSQL databases
– Theoretical issues in MapReduce
– The Scala programming language
– The Spark platform: basic and advanced issues
– Streaming, SQL, Machine Learning, GraphΧ: the basic libraries
– Data exploration using SparkR
– Algorithm design in Spark
– Graph databases
– Other systems: Giraph, GraphLab, Hama, BlinlkDB

Learning Outcomes
1. Students will get important knowledge in big data management and analytics
2. They will work in teams
3. They will be more confident by presenting their work in class
4. They will get in contact with modern big data analytics techniques with a lot of applications in Industry.

Decentralized Technologies (DWS204)

Course Content (Syllabus)
Introduction to decentralized technologies. Classification of decentralized ledger technologies. Emphasis on privacy and security aspects of decentralization. Introduction to Blockchain Technology. Introduction to Bitcoin. Scripting, Etherium, Hyperledger implementations.

Learning Outcomes
Understanding decentralized Systems principles, identify core issues involved in Internet decentralization, the models od information representation on blockchains. -based applications and Web transactions. Training on designing and developing of blockchain centered applications with special focus on the economic models as use cases. Emphasis on applications which are relevant with blockchain analysis. Emphasis on hands-on and co-working experience.

Advanced Topics in Machine Learning (DWS205)

Course Content (Syllabus)
Cost-sensitive learning, class imbalance, multi-label learning, multi-instance learning, active learning, relational data mining, interpretable machine learning, evaluating and writing scientific publications, applications of machine learning.

Learning Outcomes
Students will acquire: (a) specialized knowledge to address issues that arise in real-world applications (class imbalance, unequal classification error costs, limited training data, and data with multiple labels, instances and relations), (b) useful skills for researchers and practitioners (writing and evaluating scientific publications, applying machine learning techniques in practice), and (c) specialized knowledge about modern popular application case studies from the retail, energy and publishing industries.

Advanced Topics in Databases (DWS206)

Course Content (Syllabus)
Spatial and Multimedia Databases with emphasis on high dimensional data, Spatial Data Models, Spatial Query Languages, Spatial Storage and Indexing, Spatial Query Processing Algorithms and Optimization, Spatial Networks, Information Retrieval, Content-based Information Systems, Multidimensional Indexing Techniques, B-Trees and their variants, persistent trees, buffer trees, R-Trees and their variants, X-Trees, M-Trees, Slim-Trees, Similarity Query Processing Algorithms in Multidimensional Spaces, Similarity Query Processing Algorithms in Metric Spaces, Preference Queries (top-k, skylines), hash indexing techniques.

Learning Outcomes
After the course completion students will have a concrete view of the methods used for spatial and multimedia data management. Moreover, they will be able to apply indexing schemes (as well as other methods) in multimedia data. In addition, the assignments will help students to gain additional knowledge and solve practical problems.

Knowledge Graphs and Ontology Engineering (New)

Course Content (Syllabus)
Ontology development methodologies (Ontology Engineering), Advanced ontology design principles (Ontology Design Patterns), Knowledge Graph construction and validation (RDB2RDF, Shape Constraints and Expressions), Accessing and using semantic networks and graphs (DBpedia, YAGO, Wikidata, ConceptNet, BabelNet), Semantic repositories (RDF triplestores), RDF graph visualization (VOWL, OntoGraph, Gruff), Research trends and directions relevant to Knowledge Graphs (Natural Language Processing, Explainability).

Learning Outcomes
Students will become familiar with Ontology Engineering techniques and advanced ontology modelling concepts. They will understand the basic differences between relational databases and triplestores, acquiring skills in developing RDF-based Knowledge Graphs and mapping relational databases to RDF. The use of existing libraries and tools for visualizing, validating and querying RDF graphs will help them relate methodologies and techniques to open research directions and to a range of practical applications