Big Data Analytics

EEBDA big data analytics exercise project

Case Study 1: Descriptive and diagnostic analytics

Auditing, detection of duplicates

• Data preperation: import, conversion, filter, merge
• Data analytics with Chisquare, p-value,
• Checking equal distribution and Benford distribution

Case Study 2: Predictive Analytics with machine learning algorithms

Fraud and error detection using statistical methods of non-parametric classification.

• Machine Learning / Data Mining
• Classification algorithms:
  • K-nearest neighbors
  • Decision trees
  • Support-vector-machines
• training - predicting - evaluating

Case Study 3: Predictive Analytics using Regression while Incorporating Heterogeneity

Business Valuation M&A by means of simple multiples

• Discounted Cash Flow model and valuation using multiples
• Regression analyses: SLR, MLR, SUR
• Hypotheses test using T-test, F-test and Chow-test
• Dealing with unbalaced data
• Cross validation

Case Study 4: Natural Language Processing (NLP)

Company Perception based on unstructured data

• Tokenization, Stemming and Lemmatization, Word2Vec
• Preparation of text data for machine analysis
• Model development and evaluation of neural networks
  • torch
  • farm