Contents
1BM110 Lecture notes...........................................................................................................................2
Lecture 1................................................................................................................................................2
Lecture 2................................................................................................................................................3
Data preprocessing............................................................................................................................3
Lecture 3 Guest lecture.........................................................................................................................4
Lecture 4................................................................................................................................................6
Reinforcement learning.....................................................................................................................6
Unsupervised learning.......................................................................................................................6
Supervised learning...........................................................................................................................8
Experimental setup............................................................................................................................8
Lecture 5 Supervised learning................................................................................................................9
Regression.........................................................................................................................................9
Classification models.......................................................................................................................10
Ensemble methods..........................................................................................................................11
Performance measurement.............................................................................................................12
Lecture 7 Guest lecture........................................................................................................................13
Guidelines & methods.....................................................................................................................13
Deep learning...................................................................................................................................14
Amber car........................................................................................................................................15
Lessons learned...............................................................................................................................15
Lecture 8 Text mining..........................................................................................................................16
CRISP-DM.........................................................................................................................................16
Modelling.........................................................................................................................................17
, 1BM110 Lecture notes
Lecture 1
Conventional decision support -> emphasis on deduction:
Premise: Every swan I have seen is white.
Conclusion: All swans are white.
BI-> emphasis on induction:
premise A:all men are mortal
premise B: Pete is a men
Conclusion: Pete is mortal.
BI: data-driven decision support
emphasis on induction
Data-> model -> decision
Business/data analytics: degree of intelligence
Descriptive analytics: use data to understand the past and current performance. What is
going on, what has happened using the data collected.
o Reporting, dashboards, summarization, visualization
o Segmentation: clustering
Predictive analytics: analyse the past performance in order to predict the future. What will
occur?
o Regression & classification
o Associate rule
o Text mining: unstructured data
Prescriptive analytics: what should occur?
o Optimization techniques
o Mathematical optimization models: heuristics
Maturity & ambition level matrix
1BM110 Lecture notes...........................................................................................................................2
Lecture 1................................................................................................................................................2
Lecture 2................................................................................................................................................3
Data preprocessing............................................................................................................................3
Lecture 3 Guest lecture.........................................................................................................................4
Lecture 4................................................................................................................................................6
Reinforcement learning.....................................................................................................................6
Unsupervised learning.......................................................................................................................6
Supervised learning...........................................................................................................................8
Experimental setup............................................................................................................................8
Lecture 5 Supervised learning................................................................................................................9
Regression.........................................................................................................................................9
Classification models.......................................................................................................................10
Ensemble methods..........................................................................................................................11
Performance measurement.............................................................................................................12
Lecture 7 Guest lecture........................................................................................................................13
Guidelines & methods.....................................................................................................................13
Deep learning...................................................................................................................................14
Amber car........................................................................................................................................15
Lessons learned...............................................................................................................................15
Lecture 8 Text mining..........................................................................................................................16
CRISP-DM.........................................................................................................................................16
Modelling.........................................................................................................................................17
, 1BM110 Lecture notes
Lecture 1
Conventional decision support -> emphasis on deduction:
Premise: Every swan I have seen is white.
Conclusion: All swans are white.
BI-> emphasis on induction:
premise A:all men are mortal
premise B: Pete is a men
Conclusion: Pete is mortal.
BI: data-driven decision support
emphasis on induction
Data-> model -> decision
Business/data analytics: degree of intelligence
Descriptive analytics: use data to understand the past and current performance. What is
going on, what has happened using the data collected.
o Reporting, dashboards, summarization, visualization
o Segmentation: clustering
Predictive analytics: analyse the past performance in order to predict the future. What will
occur?
o Regression & classification
o Associate rule
o Text mining: unstructured data
Prescriptive analytics: what should occur?
o Optimization techniques
o Mathematical optimization models: heuristics
Maturity & ambition level matrix
