NETWORK SCIENCE FOR DATA ANALYTICS

2023-I

Instructors: J. Alfonso Melendez, Ph.D. e-mail: [email protected]

Fabio A. Gonzalez, Ph.D. e-mail: [email protected]
L. Fernando Niño V., Ph.D. e-mail: [email protected]

1. GOAL

To study the fundamentals of Network Science, including both commonly used and state-of-the-art
methods for data analytics, and their applications in problem solving.

Upon successful completion of the course, the student will be able to actively participate in data
analysis projects that involve networks with a data science approach.

2. COURSE LEARNING OBJECTIVES

Upon successful completion of the course, the student will be able to actively participate in data
analysis projects that involve networks with a data science approach.

Specifically, the student will be able to:

1. Understand and apply network models to complex problems.

2. Apply the modeling and analysis tools of Network Science

3. Understand the advantages and disadvantages of network modeling

4. Build, adapt and analyze a model of a complex network

1
3. CONTENTS

2
BASIC CONTENTS DETAILED CONTENTS
1. Introduction to Data Science 1.0 Course structure and evaluation.
1.1 Data Science - general overview
1.2 Data everywhere and all the time
1.3 Data Life Cycle
2.1 Social networks
2.2 Communication networks
2. Introduction to Network Science, 2.3 Transportation Networks
Basic Concepts of Graph Theory 2.4 Biological Networks
2.5 Clustering of nodes
2.6 Paths and distances
2.7 Connectivity and connected Components
3.1 Basic Definitions
3. Local and Global measures of a network 3.2 Density and sparseness
3.3 Subgraphs
3.4 Directed graphs and Multigraphs
4.1 Random Networks
4.2 Small World Networks
4. Network Models
4.3 Configuration Models
4.4 Preferential-Attachment networks
5.1 Ideas, Information, Influence
5. Dynamic Networks (Diffusion) 5.2 Threshold Models
5.3 Epidemic Spreading
6.1 Percolation Theory
6.2 Attack Tolerance
6. Dynamic Networks (Robustness)
6.3 Cascading Failures
6.4 Modeling Cascading Failures
7.1 Basic Definitions
7.2 Community Detection
7. Communities 7.3 Basic Algorithms for finding communities in a
network
7.4 Louvain Communities
8.1 Classic Node Clasification
8. Node prediction 8.2 Node similarity
8.3 Node Embeddings
9.1 Preferential Attachment
9.2 Common Neighbour
9. Link Prediction 9.3 Adamic-Adar Method
9.4 Resource Allocarion
9.5 Hierarchical Random Graphs
10. Machine Learning with Graphs (oriented to 10.1 Introduction to Machine Learning
classic Machine Learning techniques)
10.2 Node prediction
10.3 Link Prediction
10.4 Graph Prediction
10.5 Clustering in Graphs

11.1 Introduction to Deep Learning

11. Deep Learning with Graphs (New Techniques 11.2 Graph Representation learning (node
based on Deep Learning) embeddings)
3
11.3 Graph Neural Networks
11.4 Applications
3. STUDENT CONDUCT CODE

Everybody in the class is expected to show courtesy, civility, and respect for one another.
Students are expected to do their own work. Cheating, plagiarism and any other form of
academic dishonesty will not be tolerated. This will be a zero-tolerance policy. However,
students are encouraged to discuss ideas and techniques broadly with other class members.

In the assignments to be developed in groups, not doing your fair share of work or allowing
other students to receive credit for an assignment in which they did not work on is considered
an act of academic dishonesty.

It is permissible to use software and materials available from other sources as long as: 1) You
acknowledge explicitly which aspects of your assignment were taken from other sources and
what those sources are. 2) The materials are freely and legally available.

4. CLASS ATTENDANCE AND ABSENCES

This course assumes an interactive approach in its structure and in its presentation, which
requires engaged participation from all members of the class. Your presence is essential to the
success of this course and accordingly to your individual success in it. Therefore, regular
attendance is expected and considered mandatory.

Each student may have up to three absences in class during the entire semester without direct
penalization to their grade. Any unjustified absence of the student in the class could include
penalties that may result from absences in class activities, handing in take-home assignments or
deliveries of the class project.

5. COMMUNICATION CHANNELS

The following channels will be used to facilitate the communication among the class
participants. Also, all the participants will be asked to respect the purpose and rules of each one
of these channels.

● Google Classroom: This platform will be used as the main Learning Management System (LMS)
and will be the general channel to post and submit the assignments. In addition, it will be used
as a communication tool. This will be a direct channel to give support and to receive feedback
from the students. Students are encouraged to post questions and to help answer your
classmates’ inquiries via Google classroom.to administer tests and workshops.
● Course web page: It will be used to post the class material. https://tinyurl.com/224m2fry
● Github: It will be used as the suggested file repository for your term project.

6. METHODOLOGY

4
 Methodological issues of the class will include:

• Project-based learning. Students will develop a group project, preferably interdisciplinary, in

which concepts and techniques studied during the course will be applied. Each deliverable of
the project include a written component as well as an oral presentation.

• Concepts and methods will be developed with the support of readings prior to their
discussion in class.

• Presentation of some concepts and methods by the teacher to resolve doubts of the students
and reinforce the concepts studied.

• Some topics will be developed with the support of presentations by students.

• Individual and collaborative written exercises will be developed to reinforce some of the
concepts.

• Practical exercises will be developed to reinforce and apply the concepts studied in class.

7. GRADING POLICY

Activity Value

Labs: Colab Labs 1, 2, 3; each one is worth 10% 30%

Project 1: Report Research Article 50% 30%

Networkx Code 50%

Project 2: Proposal 20% 40%

Project Milestone (Prototype) 20%

Final Report 50%

Poster Presentation 10%

Extra Credit: Forum Participation, Code Add points to your overall grade, without
Contribution, etc. exceeding maximum grade.

5
* Important note: to meet the objectives of the course, the credit for partial deliveries on the project
development will only be taken into account when the project final report is delivered. To make it
clear, if the final delivery of the project does not occur, the total grade assigned to this course activity
will be zero.

8. BASIC REFERENCES

● Berthold, Borgelt, Höppner, et al. Guide to Intelligent Data Science. Springer. 2020.

● Duncan Watts. Small Worlds. Princeton University Press. 2004.

● Ryan Light, James Moody. The Handbook of Social Networks . Oxford University Press. 2020

● Ted Lewis. Network Science, Theory and Applications. John Wiley & Sons. 2009.

● Alessandro Vespignani. Dynamical Processes on Complex Networks. Cambridge University Press.

2008.

● Michelle Coscia. Atlas for the Network Scientist. tufte-latex.googlecode.com. 2021

● Vito Latora et al. Complex Networks Principles and Applications. Cambridge University Press. 2017

● Maarten Van Steen. Graph Theory and Complex Networks. Published by Maarten Van Steen. 2010

● Ernesto Estrada, Philip Knight. A first Course in Network Theory. Oxford University Press. 2015

● Stefan Thurner et al. The Theory of Complex Systems. Oxford University Press. 2018.

● Stanley Wasserman,et al. Social Network Analysis. Cambridge University Press. 1998.

● Murtaza Haider. Getting Started with Data Science: Making Sense of Data with Analytics. 1st.
Edition. IBM Press. 2015.

● J. Leskovec, A. Rajaraman, and D. Ullman. Mining of Massive Dataset. Cambridge University Press.
2010

● G.J. Myatt and W.P. Johnson. Making Sense of Data I: A Practical Guide to Exploratory Data Analysis
and Data Mining. 2nd Edition. John Wiley & Sons. 2014.

● Joel Grus. Data Science from Scratch. O'Reilly Media, Inc. 2015.

● T. Hastie, R. Tibshirani, J. Friedman. The Elements of Statistical Learning. Springer. 2001.

● William L. Hamilton. Graph Representation Learning. McGill University. 2020.

● Menczer Filippo. A first Course in Network Science. Cambridge University Press. 2020.

6
 ● Barabasi, Albert. Network Science. Cambridge University Press. 2016.

● Newman, Mark. Networks an Introduction. Oxford University Press. 2018.

● D. Easley and J. Kleinberg. Networks, Crowds and Markets. Cambridge Univ Press. 2010.

ADDITIONAL MATERIAL AND RESOURCES

Additional material and resource will be available on the web pages of the course.