– Boosting Classifier
– Bayesian Classifier
– Random Forest Classifier
– iterative Random Forest Classifier
– Bayesian Additive Regression Tree Classifier and Predictor
– GBM Classifier and Predictor
– Gradient Descent Classifier and Predictor
– Support Vector Machine
– KMEANS Variable Selection
– Convolutional Neural Network (2+3)
– Convolutional Neural Network (6+3)
– Hierarchical Recurrent Neural Network
– Lasso Ridge Regression Classifier or Predictor
Sample GitHub: https://github.com/yiqiao-yin/YinsLibrary
Title: Statistical Machine Learning Packages
Author: Yiqiao Yin
Maintainer: Yiqiao Yin
Description: This package designs a series of machine learning algorithms for data scientist.
Please contact me for product update after you made the purchase. Simply use the form below to send me an email.
This is Yiqiao Yin’s statistical software package. The entire package is written in R. The package contains all statistical machine learning functions written. One can simply open up RStudio and install this package for future use.
This product took me over 300 hours of workload. By charging average salary as that of Data Scientist in NYC ($100k which is $50 an hour), this product is charged $14,999 (300 hours by 50 dollars per hour).
Application of this statistical software package includes but not limited to:
– Predict Credit Card Default.
American credit card holders pay up to $1 trillion USD interest as of 2019 $1 trillion USD interest as of 2019 $1 trillion USD interest as of 2019. The banks such as Wells Fargo and JP Morgan Chase or Visa and MasterCard are the audience that may be of great interest of this product. If implemented, we can potentially save this country up to a trillion dollars!
– Medical Image Investigation and Classification.
The statistical software contains codes target to medical images. Using Keras package, this software package delivers advanced and complex layers of neural network architecture that includes functions such as Deep Neural Network, Convolutional Neural Network, Recurrent Neural Network, Hierarchical Recurrent Neural Network and so on. The package works on any form of medical images such as MRI or CT Scan.
– Diagnose Hidden Sequential Secret of Data Set with Insufficient Samples.
The product serves as essential toolkit to diagnose and investigate big data (Breast Cancer Data Set) problem when facing insufficient sample size and large-scale DNA sequence; integrated a dimension reduction algorithm that detects deep influential features with machine learning algorithm such as decision trees, Random Forests, Gradient Boosting Machine, or Convolutional Neural
Network (CNN, AlexNet, VGG-16, etc.) to deliver a solution with prediction accuracy of 92%.
– Data Set with Time Space Continuity
The software helped me lead the core coding task in VisionZero Project jointly with Department of Transportation to analyze, assess, and predict the number of fatalities and serious injuries given road condition using a multi-layer neural network architecture; project presents 94% accuracy at predicting serious injuries, a 70% error reduction comparing to industry peers.