Good And Bad classification of Cabbage and Potato Curry

Data Description for the Project: Good and Bad Classification of Cabbage and Potato Curry The dataset consists of 1000 samples of Cabbage and Potato Curry, divided equally into two categories: Good (500 samples) and Bad (500 samples). Each sample has been evaluated based on sensory, physical, and chemical parameters that determine its quality and acceptability. The goal of the project is to classify the samples accurately into "Good" or "Bad" based on these parameters. Below is a detailed description of the dataset: 1. Sensory Attributes These parameters assess the overall sensory quality of the curry: Appearance: Color consistency and visual appeal (measured on a 1–10 scale). Texture: Softness of the vegetables, uniformity, and absence of undesirable textures (1–10 scale). Aroma: Freshness and pleasantness of the aroma (1–10 scale). Taste: Flavor profile, including saltiness, sweetness, and bitterness (1–10 scale). 2. Physical Attributes These parameters include measurable physical properties: Moisture Content (%): The percentage of moisture in the sample. Particle Size Distribution: The uniformity of cabbage and potato pieces. Foreign Matter: Presence of impurities such as stones or extraneous particles (binary: 0 for absent, 1 for present). 3. Chemical Attributes These attributes assess the chemical quality and shelf-life indicators: pH: Acidity/alkalinity of the curry. Acidity (%): Total titratable acidity to check for spoilage. Peroxide Value: Indicator of fat oxidation (measured in meq/kg). Microbial Load: Total Plate Count (TPC) measured in CFU/g to assess microbial contamination. 4. Classification Labels Each sample is labeled as either: Good (1): Samples that meet quality standards for appearance, texture, aroma, and taste, with acceptable physical and chemical properties. Bad (0): Samples with substandard sensory, physical, or chemical qualities, or those with microbial spoilage. Dataset Characteristics Size: 1000 samples (500 Good, 500 Bad). Balance: The dataset is balanced, ensuring equal representation of both classes. Format: Tabular format with rows representing individual samples and columns representing attributes and labels. Applications: The dataset can be used for training machine learning models, such as logistic regression, SVM, decision trees, or neural networks, to classify curry quality. Significance This dataset provides a comprehensive framework to understand the quality determinants of Cabbage and Potato Curry. The classification outcomes can help improve quality control processes and ensure consumer safety and satisfaction. This concise description remains within the 3000-character limit while covering all critical aspects of the dataset. Let me know if you'd like to add any specific details or adjust the focus!本数据集为卷心菜土豆咖喱品质优劣分类项目的数据说明。本数据集共包含1000份卷心菜土豆咖喱样本，按类别平均划分为两组：优质（Good）样本（500份）与劣质（Bad）样本（500份）。所有样本均基于决定其品质与可接受性的感官（Sensory）、物理（Physical）及化学（Chemical）参数完成评估。本项目的目标为基于上述参数，将样本准确分类为“优质”或“劣质”。以下为数据集详细说明： 1. **感官属性（Sensory Attributes）** 此类参数用于评估咖喱的整体感官品质： - 外观（Appearance）：色泽一致性与视觉吸引力，采用1~10分制量化评分。 - 质地（Texture）：蔬菜的软硬度、均匀度及无不良质感情况，采用1~10分制量化评分。 - 香气（Aroma）：香气的清新度与愉悦感，采用1~10分制量化评分。 - 风味（Taste）：包括咸度、甜度及苦味在内的风味表现，采用1~10分制量化评分。 2. **物理属性（Physical Attributes）** 此类参数涵盖可量化的物理特性： - 水分含量（%）：样本中的水分百分比占比。 - 粒径分布（Particle Size Distribution）：卷心菜与土豆块的均匀程度。 - 外来杂质（Foreign Matter）：是否存在石块或无关颗粒等杂质，采用二元标签：0代表未检出，1代表检出。 3. **化学属性（Chemical Attributes）** 此类参数用于评估化学品质与货架期相关指标： - pH值：咖喱的酸碱度。 - 总酸度（%）：通过总滴定酸度检测腐败情况。 - 过氧化值（Peroxide Value）：脂肪氧化程度的指示指标，单位为meq/kg。 - 微生物负荷（Microbial Load）：平板菌落总数（Total Plate Count, TPC），以菌落形成单位每克（Colony Forming Unit per gram, CFU/g）为单位，用于评估微生物污染程度。 4. **分类标签** 每份样本均被标注为以下两类之一： - 优质（Good，标签值1）：符合外观、质地、香气与风味的品质标准，且物理及化学属性均达标的样本。 - 劣质（Bad，标签值0）：感官、物理或化学品质不达标，或存在微生物腐败情况的样本。 **数据集特征** - 样本量：1000份（优质、劣质各500份）。 - 平衡性：数据集类别分布均衡，确保两类样本的占比一致。 - 格式：采用表格格式，行代表单个样本，列代表属性与标签。 - 应用场景：可用于训练机器学习模型，例如逻辑回归（Logistic Regression）、支持向量机（Support Vector Machine, SVM）、决策树（Decision Tree）或神经网络（Neural Network），以实现咖喱品质的分类任务。 **研究意义** 本数据集为探究卷心菜土豆咖喱的品质决定因素提供了全面的分析框架，其分类结果可助力优化品质管控流程，保障消费者安全与满意度。 本说明简洁明了，字符数控制在3000以内，涵盖了数据集的所有关键要素。如需添加特定细节或调整侧重点，请随时告知。