Gender Classification Using Twitter Data

This dataset is an expansion of the Twitter User Gender Classification dataset, which is freely available on Kaggle. The aim of this data for research is to predict user gender based on textual data available on Twitter. The original dataset contained 12,894 distinct male and female twitter users with one tweet each. This was significantly expanded to 269,108 tweets by the same 12,894 users where each user had multiple tweets. Expansion method was using Tweepy to access the Twitter API. The uploaded files contains the Train and Test split used for the experiment. It contains the following: user_id - a unique id for each user gender - male or female gender:confidence - a float representing confidence in the provided gender (1 for 100%) created_at - date and time when the tweet was created tweet_id - the unique id of the text of a random tweet by the users Attached also is a simple script on Jupyter Notebook using Tweepy. This is built to retrieve a tweet’s complete information using its ID which is known as the hydration of a tweet ID. Some sample tweet id's are already in the script for testing purposes.

本数据集是对可在Kaggle平台免费获取的Twitter用户性别分类数据集的拓展。本数据集的研究目标为基于Twitter平台公开的文本数据预测用户性别。原始数据集包含12894名明确标注性别的Twitter用户，每位用户仅对应一条推文。本次拓展通过Tweepy库调用Twitter API，将样本量大幅提升至269108条推文，所有推文均来自上述12894名用户，且每位用户可对应多条推文。

本次上传的文件包含本次实验所用的训练集与测试集划分数据，其包含以下字段：
- user_id：每位用户的唯一标识符
- gender：用户性别，取值为男或女
- gender:confidence：表示标注性别置信度的浮点型数值，取值为1时代表100%置信
- created_at：推文创建的日期与时间
- tweet_id：用户随机选取的某条推文的唯一标识符，对应该推文的文本内容

附件还包含一份基于Jupyter Notebook的简易脚本，该脚本采用Tweepy库开发，用于通过推文ID获取推文完整信息，该过程也被称为推文ID水合。脚本中已预置部分示例推文ID，可供测试使用。

搜集汇总

背景与挑战

背景概述

该数据集是Twitter用户性别分类的扩展版本，包含12,894名用户的269,108条推文，用于基于文本内容预测用户性别。数据集提供了用户ID、性别、推文时间等关键字段，并附有推文ID提取脚本。