alexwolf22/ML-PANDAS-Project

A project I am doing for fun incorporating PANDAS Data Frame merging, Tidy Data Cleaning, and ML from cvs files I produced of Company CRM Info

ML-PANDAS-Project

This is a project I am doing for the fun of it to practice PANDAS dataframe aggregation, generating csv data, Tidy data Guidelines, and creating Machine Learning Models from Data and More

I created the database myself by generating CSV files; The data is generated randomly, some with a guassian distribution and priority of selection.

You can see how the data was generated in the pdf 'DBsummary'

The classification I am trying to determine is whether a customer is a good lead or not. The customer 'GOOD_LEAD' label will be determine by Data afte Exploratory Data Analysis. I catered the generation of it to make a working ML model by

* Assigning 'TEXT' data from a conversation to have only postive or negitive words with stopwords
* Making the 'CONVO_COUNT' higher for platform 'A', then 'B'', then 'C'
* Setting the convo 'LENGTH_MINS' to be generated with a guassion distribution; Mean=5, std=5
* Setting half the are code from the phone numbers to be American
* More

The goal of this project is to get more comfortable with PANDAS dataframe manipulation/transformation, EDA analysis, make a ML accurate model form 'real' data, and to have some coding fun =).

Database Schema

File Organization

Cleaning the Data

- tidyCSVtables.py

Exploratory Data Analysis

- analyzeCustomerData.py
- EDA Graphs/

Code to make the CSV Database and unique keys

- makeDataFunctions/

fullCSVtables

- Single CSV files for each table
-Have test data and will have full data once read to make full model

storedData

- Where unique/primary key Data is stored