Telco Analytics

Step I - Project Presentation

Telco Analytics

Purpose

Investment optimization and generation of new revenues to support Digital Transformation based on Data

Project Keys

 1- Semantic Data Integration
 2- Analytics - Correlação de Dados
 3- Advanced Anomaly Detection
 4- Embedded IoT Analytics to the Edge
 5- Machine Learning
 6- Open Data - For Peoples (PL 53/2018)
 7- Decision Management
 8- Immersive User Experience
 9- Geospatial And Location Intelligence
10- Digital Twins

Presentation

Link to the presentation [1]

Step 2 - Studies

Usable Books

Open this Drive Link to find 3 books [2] :
   * Data Mining for the masses 2nd edition
   * Python for data Analysis
   * Data Mining concepts and techniques

Methodology

CRISP-DM stands for cross-industry process for data mining.

The CRISP-DM methodology provides a structured approach to planning a data mining project.
Phase of the process:
     1- Business understanding
     2- Data understanding
     3- Data preparation
     4- Modeling
     5- Evaluation
     6- Deployment

Step 3 - Business Case Example

Benefits to anyone who offers this solution

   Offer a machine learning Template for data mining in order to:
     * Improve the broadband customer experience
     * Remote data management and processing with IoT optimizing investments
     * Network scanning for monitoring, testing and predicting events.

Benefits to the user

   Make Intelligent predictions more faster.

Business Models

    We can use some Classification models Like:
       - Random Forest
       - Decision Tree
       - XGBoost

Step 4 - Business-oriented prototype

Scoop

  Join Statistics and programming to better understand huge Data and make decisions in order to improve the company.

Technical details

Tools for Studing Steps:
  * Knime
  * Anaconda / Jupyter notebook : Python users
  * R studio : R users

Tools for Collaborative Work:
  * Data Lake / DW : data extraction
  * Google Data Studio : Data visualization
  * Google Colab: Python users
  * Rstudio Cloud : R users

•