e nose based Beef Quality Classification Dataset(4classes)

Electronic noses (E-noses) are increasingly used for real-time food spoilage detection and quality monitoring. Beef freshness is a critical parameter for food safety, as microbial activity produces volatile organic compounds (VOCs) that alter aroma signatures during spoilage. This dataset was developed to support the design and validation of a compact, low-cost E-nose system. The system integrates eight MOS gas sensors (MQ2, MQ3, MQ4, MQ5, MQ135, MQ136, MQ137, MQ138) and a DHT11 sensor for temperature and humidity monitoring, all connected to an ESP32-WROOM microcontroller. Each sensor was linked to an analog input pin, and raw analog-to-digital converter (ADC) values were wirelessly transmitted to a web server (iotwebserver.com). The server stored up to 50 entries before overwriting, and complete datasets were preserved by periodically downloading and merging the readings. Fresh beef samples were obtained from a local market immediately after cutting (around 6:00 AM) to ensure maximum freshness. Spoilage progression was continuously monitored for 1735 minutes (~29 hours), with sensor readings recorded every 5 seconds. Each sample was placed in a sealed sampling chamber to capture VOC emissions dynamically over time. The dataset is stored in CSV format, with each row representing one sensor reading and columns structured as follows: Minute: Elapsed time since the start of monitoring TVC: Total Viable Count (log10 scale) as a microbial reference Label: Meat quality category Class 1: Excellent (TVC < 3) Class 2: Good (3 ≤ TVC < 4) Class 3: Acceptable (4 ≤ TVC < 5) Class 4: Spoiled (TVC ≥ 5) MQ2–MQ138: Resistance values from the eight MOS gas sensors Temperature & Humidity: Recorded using the DHT11 sensor This dataset provides high-resolution, time-resolved spoilage data that can be used for time-series analysis, feature extraction, and machine learning model development for intelligent beef freshness prediction and food safety monitoring.