Data used to develop #Polar scores

We present a new approach to measuring political polarization, including a novel algorithm and open source Python code, which leverages Twitter content to produce measures of polarization for both users and hashtags. #Polar scores provide advantages over existing measures because they (1) can be calculated throughout the legislative cycle, (2) allow for easy differentiation between users with similar scores, (3) are chamber-agnostic, and (4) are a generic approach that can be applied beyond the U.S. Congress. #Polar scores leverage available information such as party labels, word frequency, and hashtags to create an accessible, straightforward algorithm for estimating polarity using text. (from the paper: Hemphill, L., Culotta, A., and Heston, M. (forthcoming) #Polar Scores: Measuring partisanship using social media content. <i>Journal of Information Technology &amp; Politics</i>.)<br>The dataset contains one plain text TSV file with the following information for each of the 55,244 tweets used to develop #Polar scores : tweet_id, created_at, user_id, screen_name, tag, shortid, sex, party, state, chamber, name. The file contains one row per hashtag, and therefore tweets may appear more than once. The Python code for calculating #Polar scores is available here: http://doi.org/10.5281/zenodo.53888

本研究提出一种全新的政治极化测量方法，包含一项新颖的算法与开源Python代码，该方法依托Twitter（推特）内容，可生成针对用户与话题标签（hashtag）的极化程度量化指标。#Polar得分（#Polar scores）相较于现有测量方法具备四大核心优势：其一，可在立法周期内全阶段计算；其二，能够精准区分得分相近的用户；其三，不受议院（chamber）属性限制；其四，属于通用型方法，可推广应用于美国国会以外的研究场景。#Polar得分整合政党标签、词频、话题标签等现有信息，构建了一套易于使用、逻辑清晰的文本极性估计算法。（引自论文：Hemphill, L., Culotta, A. 与 Heston, M.（即将出版）。#Polar得分：依托社交媒体内容测量党派倾向。*Journal of Information Technology & Politics*。）
本数据集包含一份纯文本TSV文件，共收录55,244条用于开发#Polar得分的推文，每条记录包含以下字段：推文ID（tweet_id）、创建时间（created_at）、用户ID（user_id）、屏幕名称（screen_name）、标签（tag）、短ID（shortid）、性别（sex）、政党（party）、州（state）、议院（chamber）、姓名（name）。该文件以每个话题标签为一行存储，因此单条推文可能会在文件中多次出现。
用于计算#Polar得分的Python代码可通过以下链接获取：http://doi.org/10.5281/zenodo.53888