Is there group  impact on study records?  Data mining analysis of students’ scores at Faculty of Physics, WUT

Łukasz G. Gajewski ,  Jan Chołoniewski ,  Janusz A. Hołyst 

Warsaw University of Technology, Faculty of Physics, Koszykowa 75, Warszawa 00-662, Poland


Learning is a complex cognitive process that depends not only on an individual capabilty of knowledge  absorbtion but it can be also influenced by various group interactions.  After a few years of collecting students’ records, the database at Faculty of Physics, WUT has grown enough to become an object of statistical analysis.  We have applied various data mining methods (such as PCA, clustering, Pearson product-moment correlation coefficient, mutual information, transfer entropy) for an  anonymized version of academic records in order to determine whether there are  key courses or distinctive groups of students. Moreover we have studied  whether hidden relationships and interactions among students affect the dynamics of students’ scores and if so then how strong  the effect is.


Related papers
  1. Heider balance in bilayer networks
  2. How are World Events Similar to Contagions?
  3. Hierarchical Partitions of Social Networks Between Rivaling Leaders
  4. New constant of motion for coevolving voter model
  5. Fast rumour source detection in large social networks
  6. Multiple propagation paths in locating the source of diffusion in complex networks.
  7. How much information is in New York Times articles ? 
  8. Co-evolution of the Potts model and topology of interactions
  9. Noise enhances information transfer in hierarchical networks
  10. Modeling of scientific paradigms spreading
  11. Fluctuation-dissipation relations in physiological data collected during emotional stimulation
  12. Emotional clusters observed in Internet communities
  13. Transition due to preferential cluster growth of collective emotions in online communities
  14. Sentiment and activity analysis of Manchester United football forum
  15. Collective emotions in Cyberspace, short review of Cyberemotions Project results  
  16. Online discussions modelled by an evolving Ising-like dynamics
  17. Sentiment and activity analysis of Manchester United football forum.
  18. Emotional clusters observed in Internet communities
  19.  Multidimensional analysis of emotions based on physiological signals
  20. Entropy-growth-based model of emotionally charged online dialogues
  21. Diffusion processes on coupled complex networks
  22. Analytical approach to model of scientific revolutions
  23. Stochastic resonance for information flows on hierarchical networks
  24. The Lehman Brothers Effect and Bankruptcy Cascades
  25. Observations of extreme events at Warsaw Stock Market
  26. Analysis and simulation of emotional states in Internet communities
  27. Emotional Agents at the Square Lattice
  28. Nonequilibrium phase transition due to social group isolation
  29. Scaling of human behavior in the portal browsing.
  30. Fluctuation scaling of quotation activities in the foreign exchange market
  31. Analysis of scientific productivity using maximum entropy principle and fluctuation-dissipation theorem
  32. Scaling of clusters in a one-dimensional system
  33. Scaling of human activity at several Web portal
  34. Properites of correlations in commodity markets
  35. Characteristic periodicities of collective behavior at the foreign exchange market
  36. Fluctuation scaling of quotation activities in the foreign exchange market
  37. Who wins in competition of social networks ?
  38. Networks of companies and branches in Poland
  39. Competing networks
  40. A threshold Potts model of financial markets
  41. Transition to Coherent Oscillatory Behaviour in a Route Choice Game
  42. Voter model on Sierpinski fractals
  43. How much noise is at stock markets ?

Presentation: Poster at 8 Ogólnopolskie Sympozjum "Fizyka w Ekonomii i Naukach Społecznych", by Łukasz G. Gajewski
See On-line Journal of 8 Ogólnopolskie Sympozjum "Fizyka w Ekonomii i Naukach Społecznych"

Submitted: 2015-10-11 23:35
Revised:   2015-10-12 00:25