Data Warehousing and Data Mining · Unit 8 · 5 hrs
Graph Mining and Social Network Analysis
Exam-focused notes for Graph Mining and Social Network Analysis (Data Warehousing and Data Mining, CSC420): what the TU syllabus asks and how it has actually been tested, with 5 solved past questions from this unit.
What this unit covers
- Graph mining
- Why graph mining
- Graph mining algorithm (Beam search, Inductive logic programming)
- Social network analysis
- Link mining
- Friends of friends
- Degree assortativity
- Signed network (Theory of structured balance, Theory of status, Conflict between the theory of balance and status)
- Trust in a network (Atomic propagation, Propagation of distrust, Iterative propagation)
- Predicting positive and negative links
Signed network
Define graph mining. Discuss the conflict between theory of balance and theory of status. [5]
--- Graph Mining is a subfield of data mining that focuses on discovering useful patterns, structures, and knowledge from graph-structured data. It applies data mining techniques to graphs, where data is represented as nodes (vertices) representing entities...
Full solved answer →Write down any one advantage and disadvantage of MOLAP over ROLAP. Define signed network and how do you check whether it is balanced or not? How beam search reduces the space complexity? Illustrate with an example.[10]
--- Advantage of MOLAP over ROLAP: MOLAP stores data in a multidimensional cube (pre-aggregated), which allows faster query performance. Since data is pre-computed and stored in optimized multidimensional arrays, analytical queries are answered much more qu...
Full solved answer →Show the conflict between theory of balance and status. How do you improve Apriori? [5]
--- In signed social networks, links between users can be either positive (+) or negative (-). - Balance Theory predicts the sign of a link based on the principle: "the friend of my friend is my friend." - Status Theory predicts the sign of a link based on ...
Full solved answer →Link mining
Define link mining. What are the roles of epsilon and MinPts in DBSCAN. [5]
--- Link mining is a research area that focuses on extracting useful knowledge and patterns from data that contains links or relationships between objects. Unlike traditional data mining which considers only the attributes of individual objects, link mining...
Full solved answer →Graph mining algorithm
What are application areas of graph mining? Explain the concept behind inductive logic programming with suitable demonstration. [5]
--- Graph mining deals with discovering patterns, structures, and knowledge from graph-structured data. The key application areas are: Application Area Description ------ Social Network Analysis (SNA) Investigates social structures using nodes (individual a...
Full solved answer →Make Unit 8 stick
Practice CSC420 with flashcards & quizzes