Project with Flask deployment

This machine learning project aims to analyse and find the factors related to an employee's performance and attrition. It then predicts the performance rating and the user input is taken from the webpage using flask API

Problem statement INX Future Inc, is one of the leading data analytics and automation solutions provider with over 15 years of global business presence. In recent years, the employee performance indexes are not healthy and this has become a growing concern among the top management. The CEO Mr. Brain, decided to initiate a data science project, which analyzes the current employee data and find the core underlying causes of the performance issues. He also expects a clear indicators of non-performing employees, so that any penalization of non-performing employee, if required, may not significantly affect other employee morals. The following insights are expected from this project:

This project aims to:

1. Analyze department-wise employee performance.
2. Identify the top 3 most important factors affecting employee performance.
3. Develop a machine learning model to predict employee performance based on selected features.
4. .Provide data-driven recommendations to improve overall employee performance and organizational efficiency.

Project Structure This project the following sections;

Jupyter Notebook (.ipynb)- used for exploratory data analysis, model training and visualisations. It contains the code for the trained ML model to predict employees’ performance based on the training data in the INX Employee Dataset.xls fil app.py - This contains Flask APIs that receives employee details, it computes the predicted value based on our trained model and reverts it. Employee prediction analysis.pptx - This contains the insights on the achievement of the Employee analysis performance model templates - This folder contains the HTML template to allow user to enter employee detail and displays the predicted employee salary. static - this folder contains the CSS and JS files perfomance_model.pkl - this contains the saved trained model Requirements.txt - this contains the technologies used and the libraries needed to execute and train the model successfully

Requirements ● Python (pandas, numpy, seaborn, matplotlib, scikit-learn) ● Flask (for API and web interface) ● HTML/CSS (for the frontend UI) ● Jupyter Notebook (for exploratory data analysis) ● Joblib (for model serialization)

Methodology ➢ Data Cleaning & Preprocessing ➢ Exploratory Data Analysis (EDA) ➢ Feature Selection ➢ Model Training and Evaluation ➢ Feature Importance Extraction ➢ Deployment via Flask Web Application

Key Insights ● The top 3 most influential factors affecting employee performance were:

1. Employee Environment Satisfaction
2. Employee Last Salary Hike Percent
3. Years since last promotion ● Departments with consistently low performance were identified.

● The final machine learning model was Random Forest with an accuracy of 95%.

Deployment The trained model is deployed via a Flask web application. Users can input employee data through a simple interface and get real-time performance predictions with corresponding performance ratings.

  Running the project Run app.py using below command to start Flask API python app.py By default, flask will run on port 5000.

Navigate to URL http://localhost:5000 Enter valid values in all input boxes and hit Predict.

Recommendations Based on the analysis, the following actions are recommended: ● Address Environmental Satisfaction ● Review Salary Hike and Promotion Policies ● Promote Work-Life Balance ● Focus on Underperforming Departments & Leverage High-Performing Departments ● Use the prediction model during hiring to select high-performing candidates.