I'am a software engineer consultant with a mathematical background.
I started learning Data Sciences by 2014, using customers CRM databases to deliver insights serving sales and marketing services.
Then I decided to go indeep in Data Sciences skills to be able to deliver more value from Data Sciences technologies.
The professional projects presented here are the result of my "deep learning" of Data Sciences.
http://bit.ly/Linkedin_FBANGUI
See project description on : https://github.com/dataforcast/NLP1
AdaNet is a TensorFlow framework for fast and flexible AutoML with learning guarantees. AdaNet implements an adaptive algorithm and learns neural architecture from neural Subnetworks.
See description on https://arxiv.org/abs/1607.01097
See project description on : https://github.com/dataforcast/OC_Datascientist/tree/master/P8/README.mkd
Benchmarks issued from use of classifiers based on machine learning and use of classifiers based on artificial neuron networks are exposed.
See project description on : https://github.com/dataforcast/OC_Datascientist/tree/master/P7/README.mkd
Results issued from use of NLP algorithm, supervised and unsupervized Machine Learning algorihtms are exposed.
A set of TAGs are suggested following the process of a detailed problem description post on StackOverFlow platform.
See project description on :
https://github.com/dataforcast/OC_Datascientist/tree/master/P6/Soutenance_P6_v3/README.mkd
Results of the use of generative Machine Learing algorithms along with supervized Machine Learning algorithms are experimented in order to reveal and interpret market segments from a e-commerce web database.
See project description on :
https://github.com/dataforcast/OC_Datascientist/blob/master/P5/p5_soutenance_F-BANGUI_V2/README.mkd
Results of the use of linear Machine Learning estimators are presented in order to estimate flight delays over TRANSTATS USA government database for year 2016.
See project description on :
https://github.com/dataforcast/OC_Datascientist/blob/master/P4/soutenance/README.md
This is a simple engine based on textual data available from IMDB. Collaborative dimension of such engine is not taken into account.
A set of unsupervized machine learning algorithms are experimented and benchmarked.
See project description : https://github.com/dataforcast/OC_Datascientist/blob/master/P3/README.mkd
Exploratory analysis leads to propose a scoring algorithm in order to evaluate a recipe sanity.
See project description : https://github.com/dataforcast/OC_Datascientist/blob/master/P2/README.mkd