Data Science Project Life Cycle

6 Steps of Data Science Project Life Cycle

1. OBJECTIVE

In order to build a successful business model, its very important to first understand the business problem that the client is facing.

Problem Statement Business Understanding
Work in Progress

2. DATA COLLECTION

Data required for the project is collected from various sources, such as databases like PostgresSQL, Oracle, or MongoDB or web scraping, such as Beautiful Soup.

Oracle SQL CSVs APIs
Work in Progress

3. DATA PREPARATION

Data preparation is the most time-consuming, yet arguably the most essential step in the project life cycle. It include steps like merging data sets, EDA, treating inaccurate data, etc. A model is as accurate as input data. 

Data Preprocessing Exploratory Data Analysis Feature Engineering

4. MODELING

A data science model (or algorithm) takes the organized data as input and gives a preferred output. Types of model suitable selection depends on whether the business problem at hand.

Classification Regression Clustering Time Series

5. EVALUATION

Model evaluation technique depends on the model type. The model is evaluated and assessed on unseen data before deployment.

Confusion Matrix R-Square AUC-ROC Gain Charts KS Statistic

6. DEPLOYMENT

The last step in the data science project life cycle is model deployment. A model will only contribute to solving a actual world problem, only if it is properly deployed and is active.

Dashboards WebApps APIs
Work in Progress

Keytodatascience Logo

Connect

Subscribe

Join our email list to receive the latest updates.

© 2022 KeyToDataScience