Automatic and dynamic profiling of enterprises

204 views
104 views

Published on

0 Comments
1 Like
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
204
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
2
Comments
0
Likes
1
Embeds 0
No embeds

No notes for slide

Automatic and dynamic profiling of enterprises

  1. 1. Dissertation supervised by Prof. AnaPaula Rocha and Prof. António Castro LIACC
  2. 2. 2
  3. 3. 3
  4. 4. 4
  5. 5. Context 5
  6. 6. 6
  7. 7. 7
  8. 8. 8
  9. 9. ••• 9
  10. 10. 10
  11. 11. 11
  12. 12. 12
  13. 13. 13
  14. 14. General Financial Social• Name • Stock Price • Facebook Fans• Founded • Market Capitalisation • SocialBakers Total• Origin • Revenue Score• Employees • SocialBakers Fan• Sector Score • SocialBakers Content Score • SocialBakers Engagement Score • SocialBakers Warning Score • ReviewCentre Score • ReviewCentre Reviewers 14
  15. 15. General Financial Social• Name • Stock Price • Facebook Fans• Founded • Market Capitalisation • SocialBakers Total• Origin • Revenue Score• Employees • SocialBakers Fan• Sector Score • SocialBakers Content Score • SocialBakers Engagement Score • SocialBakers Warning Score • ReviewCentre Score • ReviewCentre Reviewers 15
  16. 16. 16
  17. 17. 17
  18. 18. Frequency C4.5 Increase Classification Tree Group Similarity IndexAutomatic Stereotype Extractor 18
  19. 19. 19
  20. 20. 20
  21. 21. 21
  22. 22. C4.5 SocialBakers GSI FIc Employees < SocialBakers Total Score > 500 000 attributes 48.9% Stock Price < Activity Sector 144 dollars SocialBakers Fan Score > 5% Activity Sector: IT, consumer goods or transportation 22
  23. 23. Classification of new Enterprises 23
  24. 24. Classification Algorithms Faster than• Multilayer Perceptron clustering• Sequential minimal optimization• k-nearest neighbors• Naive Bayes• C4.5• Random Forest Missing• Radial Basis Values Function Network 24
  25. 25. Classifiers Test Dataset Initial Dataset Test Dataset Missing MissingMissing Values No Missing Values Missing Social Values Financial Values No Missing Values Missing Social Values Financial ValuesBest Algorithm Naïve Bayes RBF Network Naïve Bayes RBF Network Naïve Bayes RBF Network Error Rate 4% 17% 2% 2% 34% 3% 25
  26. 26. 26
  27. 27. 27
  28. 28. 28
  29. 29. 29
  30. 30. 30
  31. 31. 31
  32. 32. 32
  33. 33. 33
  34. 34. A valid process wasdefined for enterprise profiling A method capable ofidentifying the clusters meaning was createdA software to automate the process was developed 34
  35. 35. Hierarchical Clustering More Enterprises More attributes 35
  36. 36. Questions?

×