Exploratory Data Analysis (EDA) using Jupyter -- 3
Ngân sách $30-250 USD
Job Description:
The key is to identify customers at risk of churn early, before they get too far down the path, and to take
preemptive action to retain them and improve the customer relationship.
This is where Advanced Analytics (Machine Learning and Deep Learning modeling) comes into play.
Using Data Science and AI approaches to predict which of the customers will be loyal and which will lapse.
Your ML model(s) will help the company’s executives to understand what makes a great or bad customer
so they can take action.
Teaching machines to predict customer behavior, and communicate which customer attributes predict
specific behaviors, allows management to help build an organizational playbook for acquiring and keeping
happy customers. For example, a client success organization can reach out before there's a problem, their
marketing department can reach new customers that are less likely to churn, and their sales organization
can bring these better customers on board.
This problem is a typical classification task. You must build Machine Learning models to predict whether
a customer will churn or not.
You are asked to perform two (2) stages of analysis, based on different distribution of data:
1- Fit your models on the original (given) dataset
2- Fit your models on the modified dataset, after applying the SMOTE technique.
The Machine Learning models to be used are:
Naïve Bayes
Logistic Regression
Random Forests
XGBoost
The metrics of performance of the chosen models should be, with huge emphasis to recall:
Accuracy
Precision
Recall
F1-Score
You should follow the standard ML workflow process while building your models:
Explanatory Data Analysis (already done!)
Data Visualization (already done!)
Data Preprocessing (Data Imputation, Feature Selection & Scaling, Encode Categorical Features)
(partially done)
Address Data Imbalance (apply the SMOTE technique)
Split the training/test datasets in the 80/20 % ratio
Algorithm Selection
Modeling Building
Modeling Evaluation
Model Tuning (Hyperparameters Tuning)
Được trao cho:
Hi I'm a Data Scientist and I've worked on many projects and I'm also familiar with both python and R for achieving these tasks, I'm confident I will get the work done and achieve great results Best Regards
9 freelancer chào giá trung bình$123 cho công việc này
I can perform EDA and other required algorithms like ensemble and other classification ML models. I have good programming skills in python. thanks
I am a PhD Candidate with a specialisation in ML/DL. I have solid experience with statistical and predictive models. I can guarantee a perfect outcome and hope that we can discuss further.
Hello there I have seen and read your project "Exploratory Data Analysis (EDA) using Jupyter -- 3" and I am very much interested to help. The skill guiding ,me to handle this work are : Statistics, Data Mining, Thêm
Hi there, I can work with your description of Exploratory Data Analysis (EDA) using Jupyter -- 3 that you have provided including SPSS Statistics, Python, Statistical Analysis, Statistics and Data Mining . I am a full Thêm
