Enriched Mandarin Social Media Dataset for Sentiment Features Problem Analysis

Sentiment analysis is a crucial text classification task that aims to extract, process, and analyze opinions, sentiments, and subjectivity within texts. This dataset is a secondary dataset from the public Chinese IT technology exchange platform website CSDN, the data originates from Baidu waimai (now called Ele. me). Each comment in the data set is re-annotated manually using sentiment labels (positive, negative, or neutral), and 13 types of labels such as word order, polysemy, cyberword, argot, irony, typing errors, context, word segmentation wrong, emojis, Chinglish or mixed Chinese and English, polyphone, omission of characters, phonetic alphabet, abbreviation are used to annotate issues that affect Chinese mandarin sentiment. These re-annotations have been validated by a Mandarin linguist. This paper proposes an enhanced system model for Mandarin sentiment analysis of social media text based on transformer encoder. Researchers and developers can utilize this dataset to train sentiment analysis models, natural language processing algorithms, machine learning, or deep learning systems to accurately identify and analyze the sentiments expressed by consumers on Mandarin social media platforms.