wedamn Goto Github PK

a-bank-term-deposit-prediction

The goal of the study was to explore several DM/ML methods in extracting relevant explanatory and predictive patterns underlying the various input variables and a two-level categorical response variable.

bike_rentals_prediction

The dataset "DC_bike_rental.csv" comprise data on bike rental in the DC area from 2011 to 2012. The data was originally published by Capital Bikeshare.The cleansed data has 10 variables of mixed categorical and numeric types. This predictive modeling effort measures and compares the performance of three models (k-NN, Regression Tree, and Neural Network) on the dataset. Hyperparameter tuning is used in each model to improve prediction accuracy.

breast_cancer_image_recognition

The objective of the task is to examine the predictive (classification) power of ConvNets on high resolution breast cancer histopathological images obtained from http://web.inf.ufpr.br/vri/breast-cancer-database (Spanhol, F., Oliveira, L. S., Petitjean, C., Heutte, L., A Dataset for Breast Cancer Histopathological Image Classification, IEEE Transactions on Biomedical Engineering (TBME), 63(7):1455-1462, 2016).

german_credit_classifier

A k-NN classifier is built on a German credit dataset with 1000 observations (customers) and 21 variables.

git_practice_wed

market-basket-analysis

The raw dataset contains 9835 transactions, with a total of 169 different items. These items belong to 10 categories at level 1, and 55 categories at level 2. Thus, the dataset contains 9,835 market baskets of 169 stock keeping units (SKUs)

odwan_ba

phasefield

PRISMS-PF: An Open-Source Phase-Field Modeling Framework

sms-message-filtering-system

The filtering system has been built to classify new SMS messages into ham (legitimate) and spam messages

sms-spam-messages-text-analytics

The dataset is a collation of SMS tagged messages that have been collected for SMS Spam research. It contains one set of SMS messages in English of 5,574 messages, tagged as being ham (legitimate) or spam. In total, there are 4,827 legitimate messages and 747 mobile spam messages. Two classifiers (logistic regression and naive bayes) have been built out of the dataset.

telco-churn-prediction

The Telco customer churn data set was used for this prediction exercise. The data had 7,043 observations and 21 variables.

tensorflow

Computation using data flow graphs for scalable machine learning

tire-choice-conjoint-analysis

The data set contained 18 profiles (combinations of tire features) along with customers ranking scores. The aim of the analysis was to ascertain which tire features were of much relevance to customers.

tire-temperature-prediction

Code calculates tire temperature distributions from inflation loads.