Good And Bad Classification Of Boiled Rice

Published: 18 September 2024| Version 1 | DOI: 10.17632/cshjdvp7x2.1
Contributors:
Chandan gope Braja gopal Gope, Tanmay Sarkar

Description

For the Good and Bad Classification of Boiled Rice project, the dataset comprises 1000 samples, evenly distributed between two classes: 500 "good" boiled rice samples and 500 "bad" boiled rice samples. The classification task aims to distinguish between the quality of boiled rice, with "good" indicating well-cooked rice and "bad" representing overcooked, undercooked, or otherwise poorly cooked rice. Dataset Composition: Total Samples: 1000 Good Rice: 500 samples Bad Rice: 500 samples Data Features: Each sample is characterized by multiple features that reflect both physical and sensory attributes of the boiled rice. These features help capture the differences in quality between the good and bad samples. The specific features may include: Moisture Content: A critical determinant of rice quality, this feature captures the water content in the boiled rice, which is expected to vary between good and bad samples. Texture Score: Measured either through machine-based or manual sensory evaluation, texture indicates firmness or stickiness. Good boiled rice typically has an ideal firmness, while bad samples may be too hard or too mushy. Color Measurement: The color of the rice, usually quantified in terms of brightness or yellowness. Overcooked rice might appear darker or yellowish, while properly boiled rice retains a whiter appearance. Aroma: The intensity of aroma could serve as a distinguishing feature, as poorly cooked rice may emit different smells, often indicating burning, overcooking, or spoilage. Cooking Time: The duration for which the rice was boiled. Too short or too long cooking times generally correlate with bad rice quality. Grain Structure: A measure of grain integrity after cooking, indicating if the grains are intact, broken, or overly sticky. Good rice samples are expected to have individual, unbroken grains, whereas bad samples may show broken or clumped grains. pH Level: pH value may influence taste and texture, which is a subtle yet useful parameter to assess the quality of boiled rice. Sensory Rating: Subjective ratings provided by testers on the overall quality of the rice. This feature can be aggregated from multiple sensory dimensions (texture, taste, aroma) and serves as a holistic quality indicator. Data Collection: The samples were obtained from controlled cooking environments, ensuring consistency in raw rice type and cooking conditions. Variability between good and bad samples was introduced intentionally by altering parameters like water-to-rice ratio, cooking time, and heat intensity. The evaluation of each sample’s quality was carried out through both objective methods (e.g., moisture analysis, texture measurement) and subjective assessments (e.g., sensory panel evaluations). Data Preprocessing: Before applying machine learning models, the dataset underwent the following preprocessing steps: Normalization/Standardization: Continuous variables such as moisture content and cooking time were normalized to ensure

Files

Categories

Biological Classification, Characterization of Food

Licence