sumguy1234 Goto Github PK

assessing_ml_model_performance

demonstrates usage of evaluation metrics for machine learning models

dealing_with_imbalanced_classes

uses oversampling on customer usage statistics to balance the data fed to the ml model

fbprophet_electricity_price_predictions

uses fbprophet with tutorial-like informative comments to demonstrate how fbprophet works

forecasting_mercado_using_prophet

use fb prophet to forecast sales as well as search trends. Also analyzed search trends to see if there was any effect on sales and stock price.

google_trends_v_stock_volatility

time series analysis of Apple from October 2015 to October 2020 - analyzes whether any predictable relationship exists between google trends data, aapl stock price, and aapl volatility/return metrics.

learning

Class files

logistic_regression_example

neural_network_fraud_identification

Usage of deep learning and optimization to identify fraudulent transactions

predicting_customer_creditworthiness

Uses a dataset of historical lending activity from a peer-to-peer lending services company to build a model that can identify the creditworthiness of borrowers.

predicting_customer_referrals

uses logistic regression and evaluative statistics on the ml model

predicting_employee_attrition

Losing employees can carry a great financial cost for companies. This app was constructed using a human resources dataset from IBM that was shared on Kaggle. The dataset contains employee profiles and attrition info such as age, department, role, work/life balance, and so forth. This data is used to generate a deep learning model that predicts whether an employee is likely to depart from the company within the next two years given their profile.

predicting_fraudulent_transactions

use logistic regression to make predictions

preventing_credit_card_defaults_pt1

Usage of neural network machine learning techniques to predict credit card default risk

pychain_blockchain

random_oversampling_effictively

Taking a look at data from the SBA (small business administration) to measure effectiveness of random oversampling. This dataset is imbalanced. Using the dataset and oversampling the model predicts which SBA loans are likely to default.

scarf_demand_forecasting

Use time series forecasting to analyze google trends data to validate market opportunities that would help the Aymara indigenous people in Bolivia export alpaca wool scarves to different countries/regions

sentimental_trading_analysis

Trading bot using twitter data provided by augmento.ai

sumguy1234

Config files for my GitHub profile.

tensorflow.keras_explainer

Explanation of neural networks using dummy data and tensorflow.keras modules specifically Sequential and Dense

test

Testing Github and Git interactivity

test2

venture_capital_funding_analyzer

This app looks at a dataset of more than 34k organizations that have received funding from a venture capital firm over the years. The csv contains a variety of info about each business including whether or not it ultimately succeeded. This app uses a deep neural network machine learning approach to create a binary classification model that will predict whether an applicant will become a successful business.