乳腺癌数据集
该数据集是一个乳腺癌诊断相关的医学数据集，包含肿瘤的多种细胞学特征（如肿块厚度、细胞大小均匀性、细胞形状均匀性、边缘粘附性、单个上皮细胞大小、裸核、平淡染色质、正常核仁和有丝分裂）以及良性或恶性分类标签，数据模态为临床表格数据，主要用于机器学习模型在乳腺癌诊断、分类预测和特征分析方面的研究与应用。
基本信息
资源简介
乳腺癌数据集
【数据集背景】
本数据集源站标题为《jrao1724/Breast-Cancer-Dataset-KNN》。源站首次发布：2020-05-16。
【数据内容】
源站原始描述：Jupyter Notebook with an implementation of the K-Nearest Neighbors machine learning algorithm to predict if a female can be diagnosed with breast cancer after considering a certain set of features. Dataset can be found in the UCI Machine Learning Repository. # A kNN Model to Predict Benign or Malignant Tumors in Patients Diagnosed With Breast Cancer This project was initially for my Statistics class, but I later revised and added features to it. The script found in the Jupyter Notebook performs…
【获取与许可】
源站页面：https://github.com/jrao1724/Breast-Cancer-Dataset-KNN
下载信息
注册下载
需要注册 Kaggle 账号并登录后下载，适合需要跟踪下载记录和使用 API 的用户。
暂未开放公开下载
无需注册即可直接获取公开样本或文档，适合快速预览和评估数据集质量。
免登录有偿下载
公开数据集受托下载与技术交付服务。
提供高速下载与技术交付服务（收技术服务费，非数据销售）
暂未开放当前数据集主要来源为 Kaggle 公开托管，完整影像包建议通过原始链接或 Kaggle API 获取。
使用方式
数据集获取
git clone https://github.com/jrao1724/Breast-Cancer-Dataset-KNN.git
curl -L -o repo.zip https://github.com/jrao1724/Breast-Cancer-Dataset-KNN/archive/refs/heads/master.zip
unzip repo.zip
源站 README 摘录（使用方式）
A kNN Model to Predict Benign or Malignant Tumors in Patients Diagnosed With Breast Cancer
This project was initially for my Statistics class, but I later revised and added features to it.
The script found in the Jupyter Notebook performs predictions for finding out whether a tumor in a patient who has been diagnosed with breast cancer is benign or malignant.
Description of The Dataset
The dataset can be found here: Breast Cancer Dataset
The features in the dataset are:
- ID Number
- Clump Thickness: 1 - 10
- Uniformity of Cell Size: 1 - 10
- Uniformity of Cell Shape: 1 - 10
- Marginal Adhesion: 1 - 10
- Single Epithelial Cell Size: 1 - 10
- Bare Nuclei: 1 - 10
- Bland Chromatin: 1 - 10
- Normal Nucleoli: 1 - 10
- Mitoses: 1 - 10
- Class: 2 - Benign, 4 - Malignant
The Code
To complete this project, I used pandas, seaborn, matplotlib, numpy, and scikit-learn modules.
I also needed to polish the data before starting my analysis of it. To do so, I converted the Class field to be either a 0 or 1, rather than the original 2/4.
I also removed all rows with NaN values, just to make the analysis easier for myself.
The Analysis
To start off the analysis, I created a Confusion Matrix using seaborn to give myself an idea of which features are the most strongly correlated.
I then wrote a small loop to find predictions using different nearest neighbor values, and appended all of these prediction percentages to a list. Using matplotlib, I plotted these prediction values and chose the nearest neighbor value associated with the highest prediction percentage.
Finally, I performed a k-Fold Cross Validation across my dataset to make sure I was not underfitting nor overfitting my model to my dataset.
Check it out the code in k-Nearest Neighbors Prediction Model of Breast Cancer Diagnosis from The University of Wisconsin.ipynb!
数据缺失？
依托陕西、四川两大基地，我们与超过十家三甲医院建立直接合作关系，覆盖合规授权、采集治理、专业标注、数据交付的全流程，为AI医疗团队提供即拿即用的高质量临床数据。
⚡️ 需要数据支持或标注服务？立即联系我们获取专业方案。
