Statistical Notes

Here are few statistical writings I have done during my academic years:

papers:

Italian:

• Master thesis : Boltzmann Machines for Topic Modeling.

• Bachelor thesis : Cluster Algorithms for Sector Analysis

• Analisi di concentrazione del settore dei trasporti : exam project for Economia e Dinamica Industriale (Master)

• Airlines delay : exam projects for Fondamenti di data analytics (Bachelor)

English:

• Milan House Prices : fourth place on a private Kaggle competition, 50 participants. Projects for the exam of Data Mining (Master)

presentations:

Italian:

• Master thesis presentation : Boltzmann Machines for Topic Modeling.

• STMicroelectronics Lab Data Challenge : Solution proposed for the Lab Data Challenge proposed by STM. The aim was to predict the presence of a person in a room from the wifi noise frequencies coming from the router. (Master)

• Proposta di Fedelity Card per clienti di profumerie: Exam project for Analisi di mercato quantitative (Master)

• Segmentazione dei consumatori del mercato vestiario: Exam project for Analisi di mercato quantitative (Master)

• Analisi di concentrazione del settore dei trasporti: exam project for Economia e Dinamica Industriale (Master)

• Segmentazione caratteriale dei manager: Exam project for Statistica Aziendale (Master). An example usage of the software KNIME.

• Alberi di decisione per l’imputazione dei dati mancanti nei questionari: Exam project for Statistica Aziendale (Master). An example usage of the software KNIME.

• PCA per l’analisi dei questionari: Exam project for Statistica Aziendale (Master). An example usage of the software KNIME.

English:

• Automated Essay Scoring : for the exam of Machine Learning (Master), based on text mining task

notebooks:

Italian:

• STMicroelectronics Lab Data Challenge : Solution proposed for the Lab Data Challenge proposed by STM. The aim was to predict the presence of a person in a room from the wifi noise frequencies coming from the router. (Master)

• Retina Cancer : first place solution on a private Kaggle competition, 8 participants, for the exam of Statistical Learning (Master)

• POS tagger : Project made for the exam of Statistical Learning (Master). It is not a good way to train a POS tagger. Maybe If You want to train a neural network You should use directly python, and don’t switch to Keras for R.

• Analisi di concentrazione del settore dei trasporti: exam project for Economia e Dinamica Industriale (Master)

• Statistica Spaziale in R: R code notes from Spatial Statistics (Master)

English:

• Kernel Clustering for text : If You try to apply common clustering algorthms to Bag of Words Text representations, You will obtain random clusters. However, when I use Kernel distances with Kmeans for the Italian Constitution articles, I get pretty good results.

• Pytorch MNIST : a very simple exercise to train a neural network with the most common python library in the most classical and simple task (written digit classification from images).