A curated dataset of multi-channel TV program schedules for optimization and benchmarking

Area of Science:

• Computer Science
• Artificial Intelligence
• Operations Research

Background:

• Smart TV program guides require efficient scheduling algorithms.
• Existing datasets may lack diversity or real-world complexity.
• Automated data collection and standardization are crucial for large-scale analysis.

Purpose of the Study:

• To present a comprehensive dataset of 515 Smart TV scheduling instances.
• To provide a standardized format for benchmarking scheduling algorithms.
• To facilitate research in optimization and constraint-based scheduling.

Main Methods:

• Data collected from IPTV EPG, EPG.PW, and YouTube Data API.
• Automated Python scripts used for data acquisition and processing.
• Instances standardized into JSON format with consistent scheduling periods, channels, program details, and quality scores.

Main Results:

• A dataset of 515 diverse Smart TV scheduling instances is now available.
• Each instance includes scheduling periods, channel information, non-overlapping program slots, categories, and quality scores.
• Instances incorporate strict availability rules and flexible time preferences.

Conclusions:

• The dataset serves as a valuable benchmark for optimization and constraint-based scheduling.
• It enables testing of algorithms under time-dependent constraints and objective formulations.
• The data facilitates feature extraction for dataset characterization and dimensionality reduction.