This repository contains code and documentation for a project aimed at enhancing the customer experience on stc TV through data analysis, predictive modeling, and recommendation systems, as part of STC's virtual work experince offered through Misk.
This project involves:
- Statistical Analysis: Conducting comprehensive statistical analysis to uncover key insights from user data.
- Viewer Categorization: Classifying and analyzing viewer categories based on program types (movies and series).
- Viewing Patterns: Investigating user viewing patterns to determine preferences for standard definition (SD) versus high definition (HD) quality.
- Predictive Modeling: Developing models to forecast customer views over the next two months, identify peak viewing periods, and pinpoint potential customers.
- Recommendation System: Building a system to suggest content based on the viewing habits of users with similar preferences, for both Arabic and English content.
- Data Storytelling: Preparing and delivering presentations showcasing analysis results and insights, offering actionable recommendations to improve service quality.
- User Experience Enhancement: Compiling findings into PowerPoint presentations to support further enhancement of the user experience.
Describe the statistical values for the given data set, including:
- Arithmetic mean
- Standard deviation
- Maximum and minimum values Classify and analyze viewer categories according to the program type (Program Class) - whether it’s a movie or a series Study the viewing patterns of users and identify the category that watches stc TV in standard definition (SD) quality versus the category that watches it in high definition (HD).
Build an easy and simple model that allows the decision-makers to predict the number of expected customer views in the next two months, determine the expected peak period, and identify the potential customers.
Recommendation System
- No output data is shared in this repository, as I do not own the data. Only the code I worked on is included.