I am a Management student currently attending Bogazici University. My data science interest started in the period when I had taken the Introduction R Programming course which I worked for it as a Teaching Assistant later. I think I am much closer to the technical side of Business Administration and I am trying to sharpen my data analyzing skills to make them fit for the requirements of technical and analytics subjects. Here you can find my LinkedIn.
The Data Governance Specialist Alyssa Columbus introduces Data Privacy & Data Governance concepts by making the definitions firstly and tells about their standards, regulations, and privacy laws made all over the world. Following, she elaborates on the Data Governance as mentioning ideas lying behind it and its mission statement. Moreover, she presents a proper framework for Data Privacy & Governance constituting 4 layers which are Executive Stakeholders, Data Privacy & Governance Leaders, Council, and Stewards from bottom to top respectively. The rest of the presentation consists of informing the audience about “How to Tag Metadata” and “How to Anonymize Data” using R. For a person like me who took recently a course about Personal Data Privacy and Data Protection, this video was very refresher and informative about how to merge my knowledge about privacy regulations and my knowledge about R programming.
CTO of Appsilon Data Science, Marek Rogala starts his presentation by telling about what his company doing. He makes introductive narratives regarding satellites, their systems, and features needed to be known while doing satellite imagery. For example, resolution types such as spatial, temporal, spectral, and trade-offs between them, resolution quality, and availability of the satellites for a period. After all, he comes to the subject of using R for satellite data and tells about strong and weak points of R at doing satellite imagery. He explains details of how they also make deep learning with the help of Keras R and neural networks in their work. Finally, he elaborates on their services in which they used satellite imagery for Agriculture, Real Estate, and Finance&Insurance.
Using Deep Learning on Satellite Imagery to Get A Business Edge
Biomedical Engineering Lecturer at Czech Technical University, Lubomir Stepanek makes an introduction about the relationship between facial emotions and human facial attractiveness and tells about how plastic surgery deals with facial emotions in the terms of their improvement. After that, he starts to explain their study which aims to identify geometric features of a face and classify the facial emotions accurately. They gather the facial image data of people before and after rhinoplasty operation and facial expression data according to the given incentives. The team derives some metrics and angles from the points defined on the facial images and uses those measurements while classifying of facial emotions via t-statistics, p-value, and data-trees. At the end of the study, they achieved to determine that some metrics on the face play a huge role to increase facial attractiveness, facial emotions categorization shows high accuracy via neural networks, and finally mouth, eyebrow, and eyes affect the accuracy of facial emotion categorization in the descending order.
Using Machine Learning Methods and R to Classify Facial Image Data For Plastic Surgery
John Blischak, a post-doctoral scholar at the University of Chicago introduces his R package workflowr which makes available to organize files, track intermediate results, and sharing the results all together in a proper and easy way. While designing this package, he combines rmarkdown, knitr, Git, and Github. workflowr creates an organized directory structure at first, then it binds each source code and its result with a version number. Additional recording and setting seed features make the package very charming in the case of increasing reproducibility. In the end, it gives the chance to the user to share his work in his Github Pages via just a function embedded in the package.
The workflowr R Package: A Framework for Reproducible and Collaborative Data Science